Ad Hoc Networks 114 (2021) 102409

Contents lists available at ScienceDirect

Ad Hoc Networks
journal homepage: www.elsevier.com/locate/adhoc

Survey paper

A comprehensive survey on LEACH-based clustering routing protocols in

Wireless Sensor Networks
Ikram Daanoune ∗, Baghdad Abdennaceur, Abdelhakim Ballouk
B.P. 146 Mohammedia, 20650, Morocco
Department of Electrical Engineering, Laboratory of Electronics, Energy, Automatics and Data Processing, Faculty of Sciences and Techniques of Mohammedia,
Hassan II University of Casablanca, Morocco

ARTICLE INFO ABSTRACT

Keywords: During the previous few years, Wireless Sensor Network (WSN) appears as an active research domain due
Wireless Sensor Network to the wide use of this technology in several applications, such as military, health, automation, and so
LEACH forth. WSN supervises physical attributes where human activity is difficult or impossible. The WSN is a
Clustering
set of multiple sensor nodes that are haphazardly deployed in a specific space. These nodes sense data and
Routing protocols
record values continuously and send these data to the Base Station (BS) through other sensor nodes. Several
Energy consumption
issues are encountered in the WSNs, including energy consumption, deployment of sensor nodes, routing
algorithms, energy efficiency, Cluster-Head (CH) selection, robustness, etc. Numerous routing protocols have
been developed by researchers to resolve these constraints, and several techniques of optimization are proposed
to define the optimal road between the transmitter and the receiving node. Hierarchical routing protocols
increase the network lifetime by enhancing its performance. The most popular hierarchical method is LEACH
(Low Energy Adaptive Clustering Hierarchy) that grouped sensor nodes into clusters. Each cluster consists of
normal nodes and is controlled by a CH, which is elected by cluster members to collect their data and send
it to the BS. The LEACH protocol decreased the energy consumption in WSNs. In this survey, we provide a
comprehensive study of LEACH descendant clustering protocols. This survey is the first study to classify LEACH-
based routing protocols into CH selection, data transmission, and both CH selection and data transmission
techniques. This survey is compared to other actual surveys. To evaluate these protocols, we look at many
metrics, such as the CH selection method, communication method, the scalability, energy efficiency, mobility,
localization of nodes, and so forth. According to these metrics, we propose a comparative analysis of these
clustering routing protocols. This survey discusses also the strengths and limitations of each LEACH-variant
protocol. Conclusively, the paper terminates with recommendations on future research fields in WSN.

1. Introduction zone; (2) detect and register the incidence of events in this supervised
area; (3) measure the appropriated criterion [2].
The last decades have seen a growing trend toward Wireless Sensor In WSN, sensor nodes are alimented by batteries (in most of the
Network (WSN) because of the significant use of the WSNs in several cases they are neither rechargeable nor replaceable). Energy is con-
domains, such as military, civilian, and industrial. WSN is a network sumed in collecting data, treatment, and transmitting or receiving
constituted by a group of micro sensor nodes deployed randomly in packets. Therefore, the energy consumption is the major constraint in
a specific area, and at least one sink (or Base Station (BS)). In WSNs, WSNs. Besides, another constraint that influences and decreases the en-
there are diverse kinds of sensor nodes: source sensor node (normal ergy efficiency of sensor nodes is the redundancy of data. Consequently,
node), intermediate sensor node (especially Cluster-Heads (CHs) in economizing energy to increase network lifetime is a challenge in WSN.
clustered networks), and BS [1]. These micro sensor nodes can monitor
In this context, many studies of WSN show the importance of routing
physical and environmental phenomenons like temperature, pollution,
protocols.
heartbeat, etc. Sensor node devices collect data and send it to the sink,
The routing protocol is a procedure of choosing the right route
then through the internet, the user can get information from the sink as
for the data to move from source to destination [3]. The objective of
shown in Fig. 1 The main objectives of WSN are: (1) control a specific

∗ Corresponding author at: Department of Electrical Engineering, Laboratory of Electronics, Energy, Automatics and Data Processing, Faculty of Sciences and
Techniques of Mohammedia, Hassan II University of Casablanca, Morocco.
E-mail address: [email protected] (I. Daanoune).

https://doi.org/10.1016/j.adhoc.2020.102409
Received 23 July 2020; Received in revised form 25 December 2020; Accepted 26 December 2020
Available online 16 January 2021
1570-8705/© 2021 Elsevier B.V. All rights reserved.
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

Fig. 1. WSN architecture.

Fig. 2. WSN applications.

routing protocols is to realize the scalability of the network, strengthen data and conserve energy. Furthermore, the length of the stored routing
data transfer, and energy efficiency [4]. During recent years, many table on each sensor node is reduced since roads are established in
routing protocols have been developed for the WSN to extend the net- clusters [9].
work lifetime through decreasing the power consumption. These pro- One of the most famous clustering routing protocols of WSN is
posed routing protocols depend on several metrics, like communication LEACH [10]. In LEACH, sensor nodes are gathered into groups named
model, network structure, and topology [5]. clusters. Each cluster selects one node to become a leader (cluster-
The literature review is based primarily on hierarchical routing in head) and the rest of the nodes become cluster members (or non-cluster
which the different techniques of the cluster-based routing protocol are heads) [11]. Every common node collects its own information from the
considered [6]. environment and sends it to its CH. The CH groups data received from
Clustering puts an efficient technique for energy balancing in WSNs cluster members and does compression, then sends aggregated data to
using data aggregation [7]. Clustering routing protocols are used in the sink.
WSN to reach energy-efficient because the clustering method decreases The LEAH protocol ameliorates the lifetime of WSNs by decreasing
the number of packets routed in the network. In WSNs, clustering the number of transmission packets via clusters formation. However, it
has numerous drawbacks, such as the random selection of CHs without
has various advantages like more scalability, less load, less energy
considering either the distance or the current energy of nodes. Another
consumption, data aggregation, collision avoidance, load balancing,
drawback is the single-hop used by CHs to communicate with the sink,
latency reduction, fault-tolerance, more robustness, a guarantee of
which makes LEACH is not applicable for large networks.
connectivity and maximizing of the network lifetime, etc.
Several surveys [12–16] have been made to examine the popular
Less energy consumption: data aggregation serves to reduce com-
LEACH protocol and its descendant clustering hierarchical algorithms
munications data and conserve energy. Besides, intra-cluster and inter-
by using different classifications based on different metrics, like the
cluster communications allow less energy consumption for the network clustering method, mobility of nodes, energy efficiency, and scalability.
since they decrease the cost of the distance between the farthest nodes The existing survey on the LEACH improvement protocols has been
and the sink. Moreover, the fact that just the CHs that responsible mostly restricted to limited comparisons of some criteria. Thus, some
for transmitting data to the sink, it also helps to conserve energy of them have not treated LEACH descendants protocols in much de-
consumption. tail. The goal of this survey is to review recent research into the
Data Aggregation: Serves to group the multiple data of nodes to LEACH based routing protocols. The present work provides a recent sur-
suppress redundant transmission and furnish aggregated data to the vey of clustering routing protocols improved of LEACH protocol. This
sink [8]. Consequently in clustering, CH aggregates the received data survey explores many LEACH variant protocols by using a novel taxon-
from cluster members, then forwards the grouped data to the sink, omy, which relies upon on kinds of metrics, such as energy-efficiency,
which allows economic energy. clustering method, and communication technique.
Less Load: In the clustering scheme, the aggregation or fusion is In clustering hierarchically based protocols, both CH selection and
effected by the CHs, and clusters members communicate with their data transmission techniques have importance in extending the life-
corresponding CHs, which help significantly to reduce transmission cycle of WSN. For this reason, this survey proposes a new taxonomy,

2
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

which classified LEACH enhancement protocols into three classes: (1)

clustering protocols based on CH selection to improve the LEACH pro-
tocol; (2) protocols based on data transmission techniques to enhance
LEACH; (3) protocols based on both CH selection and data transmission
methods. Then in each class, protocols are classified in terms of some
criteria, such as the residual energy, mobility, distance, single-hop,
multi-hop technique, and so forth.
The first-class concerns LEACH descendant protocols based on the
CH selection. This class is established in terms of some metrics, like
residual energy, energy-efficiency, mobility of nodes, distance, and cen-
tralization. The second one concerns protocols based on data transmis-
sion techniques, which groups protocols into three categories: protocols
Fig. 3. WSN Units.
that use the single hop to route their data, protocols use the multi-
hop, and protocols that use both techniques. The third class contains
protocols that have improved both CH selection and data transmission
compared to the classical LEACH algorithm. some applications require the mobility of nodes like objects tracking,
The contributions of the recent survey are as follows: intelligent transportations service, meteorology, seismic monitoring,
infrastructure protection, they called MWSN (Mobile Wireless Sensor
• LEACH descendant clustering routing protocols are classified
Network). Furthermore, when sensor nodes distributed in industrial
based on CH selection and data transmission techniques through
societies or in factories to control machines, it affected IWSN (Indus-
a new taxonomy based on numerous criteria, such as energy-
trial Wireless Sensor Network). Besides, UWSNs (Underwater Wireless
efficiency, distance, residual energy, single hop, and multi-hop.
Sensor Network) distribute sensor nodes underwater to monitor sea
• An interesting comparison between discussed protocols based
pollution, underwater navigation, catastrophe prevention, etc.[21]
on several metrics, like energy efficiency, communication tech-
niques, clustering methods, the CH selection, deployment, the
scalability, etc. is performed to facilitate the selection of the most 2.2. WSN units
effective protocol for researchers based on their needs.
• A comparison of the present survey with other previous surveys. WSN is a non-infrastructure network that consists of a set of au-
The resting part of the survey proceeds as follows: the second tonomous SNs for detecting and monitoring an area in the environment.
section offers an overview of WSN which introduces a background of These SNs congregate to built a wireless network, which is deployed
WSN, its applications, its components, and its constraints. Further, This randomly in the desired geographical region to monitor. Every SN
section indicates an overview of the conventional LEACH protocol and detects certain data and transmits it to the BS through other SNs, using
an overview of several existing surveys. The third section introduces the wireless communication.
proposed taxonomy of LEACH descendant protocols, which describes Every node in the WSN has four main units as illustrated in Fig. 3:
the routing protocols for each category and provides a comparative (1) sensing unit that including one or more sensors for data acqui-
analysis of different discussed protocols. Future works and research sition from the environment [22], then an ADC (Analogue to Digital
challenges are given in the fourth section. Finally, the final section Converter) to convert the analog signal produced by the sensors based
provides a conclusion of this survey. on the observed phenomenon to numerical signal; (2) treatment unit
consists of a processor, storage, and routing protocol. This unit executes
2. Overview of WSN the communications protocols which allow communication between
nodes; (3) wireless transmission unit involves two radio frequency
WSN offers new technology with convincing benefits in contrast modules of emission and reception that offer a wireless communication
to traditional wired networks, which has the advantage of decreasing
to connect the node to the network; (4) power unit (generally are
installation and servicing charges of monitoring systems [17].
batteries) is the unique source of energy that supplies all the others
WSN comprises a group of small, intelligent electronics devices
units.
recognized as sensor nodes (SNs) deployed over the network, and at
least a BS connected to these sensor nodes via wireless channels. The
number of SNs can vary from a few to various hundred relying on the 2.3. WSN constraints
applications [18]. Sensor nodes sense and supervise the physical and
environmental phenomenon, such as health monitoring, temperature, WSNs have numerous characteristics and they have imported sev-
humidity, etc. To reach the surveillance of a particular environment, eral benefits over the traditional wired networks. In contrast, many
sensor nodes can communicate and collaborate with others to send
constraints are faced by WSNs. In the first one, in almost WSN appli-
their gathered data to the network via the BS [19]. Fig. 1 presents the
cations, the deployment of SNs is done randomly. Another constraint
architecture of WSN.
in WSNs is the energy, due to the limited batteries power which they
are very difficult to recharge or replace them. Besides, data redundancy
2.1. WSN applications
constraints can also decrease the energy-efficiency. WSNs have also lim-
itations on processing capability, memory capability, communication
Recently, WSN has been increasingly important in the world by
its application in numerous domains, where human activities are very bandwidth, and transmitting the collected data from SNs to the BS.
difficult [20], such as military monitoring, environment, industrial, In WSN, the power unit provides energy to all SN components.
home automation, medical surveillance, underwater, etc. We grouped Furthermore, SNs waste energy in data collecting, in the treatment,
these WSN’s applications in taxonomy as shown in Fig. 2. the transfer, or reception of packets. Therefore, the SN can dissipate
When sensor networks are used in the medical and health do- its initial energy early in a short time, which influences the lifetime of
main, it concerns WBAN (Wireless Body Area Network). WSN is called the entire network. Consequently, limited power is a considerable con-
WMSN (Wireless Multimedia Sensor Network) when sensor nodes are straint in WSN. In this context, several routing protocols are appeared
used to transfer multimedia, such as images, audio, and videos. When to solve such scenarios.

3
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

Fig. 4. Set-up stage of LEACH.

Fig. 5. LEACH architecture.

2.4. Overview of the LEACH approach

To resolve the constraints posed by WSN, several routing pro- 2.4.2. Data communication stage
tocols are introduced, which serve to improve the consumption of The data communication stage is composed of numerous frames and
energy-efficiency and extend the network lifetime. The most interesting it is much longer than the first stage [26]. In this stage, each normal
protocols are the hierarchical ones, like the LEACH protocol. These node sends its collected data to its potential CH using a time slot in
protocols provide the clustering concept. In this section, firstly, we the TDMA schedule. After the fusion of data by CHs, these data will be
provide an overview of the LEACH protocol, and then in the second forwarded to the sink [27] using the CDMA code to avoid the collision.
subsection, many existing surveys are discussed. Fig. 5 illustrated a WSN protocol based on LEACH architecture,
LEACH [10] was proposed to guarantee a balanced energy utiliza- which is split into clusters.
tion and to enhance the efficiency of WSNs by partitioning the network To simulate the LEACH protocol, an algorithm is given in [28].
into multiple clusters, and through a random CHs rotation [23]. LEACH
is a MAC protocol based on the TDMA method, which is considered as 2.4.3. Advantages and disadvantages of LEACH
the most famous and the first hierarchical routing protocol used clus- To avoid interference between nodes, LEACH uses the TDMA sched-
tering. LEACH is a routing protocol based on the clustering mechanism ule. To prevent collision among CHs, the CDMA code is taken up in the
in which nodes are associated together to establish separate clusters. LEACH approach. Besides, in LEACH, nodes remain in sleep mode until
Thus, every cluster has a head node called CH that aggregates cluster their turn to send data arrives. Thus, it decreases the number of trans-
members data, then sends it to the sink [24]. mission packets through forming clusters and selects CHs. Although,
The tasks of the LEACH protocol are characterized by turns, and these advantages, LEACH stills limited by several disadvantages. The
each turn is involved in two main stages, which are the set-up and deployment of CHs and the number of CHs in each round cannot
the data transfer stage. The set-up stage includes the election of CHs, be guaranteed owing to the random selection of CHs. Thereby, in
the formation of clusters, and the creation of the TDMA schedule as LEACH, nodes with low remaining energy have the same percentage to
illustrated in Fig. 4. become CH as well as nodes with high residual energy. Consequently,
if these nodes become CHs, they will exhaust their energy early than
2.4.1. Set-up stage other nodes. Another problem of LEACH is the CHs selection ignores
The first stage is intended to select CHs and form clusters in the the remaining energy of nodes. Despite that, the CH supports higher
network. Every sensor node votes to turn into a leader (or CH) of traffic charges than normal devices due to its role which is represented
its cluster or not for the present epoch. This decision depends on in acquiring data from all its cluster members, also aggregation and
the random number choosing by the sensor node at the starting of sending to the sink [29]. Then, CHs use the single-hop to direct data to
each epoch. The choosing number is between 0 and 1. The sensor the BS that makes LEACH not adopted for a large network.
node becomes a CH in the present epoch if this number is lower
than the threshold function defined in Eq. (1)[10]. On the other hand, 2.5. Existing surveys
it becomes a normal node [25]. The selection of CHs following this
manner leads to distribute energy between all the network nodes. Authors in [12] discussed six variants of the original LEACH algo-
{ rithm, which get over the non-uniform CH distribution posed in the
𝑃𝐿
1−𝑃𝐿 ∗(𝑟 𝑚𝑜𝑑(1∕𝑃𝐿 ))
,𝑛 ∈ 𝐶 LEACH algorithm. This survey described some LEACH-based hierarchi-
𝑇 (𝑛) = (1)
0 otherwise cal routing protocols, such as LEACH-C, MM-LEACH, TL-LEACH, SEP,
V-LEACH, and MOD-LEACH then offered differences of each protocol
Where 𝑃𝐿 introduces the percentages of CHs in each epoch, r is the from the conventional LEACH protocol, based on the energy dissipa-
present epoch, and C is a set of sensor nodes that have not yet been CH tion. The drawback of this survey is that the authors did not use any
in the period 1∕𝑃𝐿 epochs. classification to examine these protocols. Thus, they just considered six
Once CHs are chosen, they inform the network nodes of their status LEACH descendants.
by broadcasting an advertisement message. Afterward, every sensor The survey in [13] studied the evolution of energy consumption
node joins its corresponding CH depending on the strength of the of several protocols derived from LEACH. This review focused on the
received signal from CHs. The acquired signal has high force if the CH CH election mechanism. To evaluate the performance of these descen-
is very near to a sensor node. Once a node chooses its join CH, this dant protocols, authors simulated and compared them to the LEACH
CH is informed by a request message to adding this node to its cluster protocol according to their energy consumption. In [13] a comparison
members. study of various hierarchical routing protocols of LEACH is done based
When clusters are established, each CH sends to its members a on certain metrics like the number of CHs and the hop count. The
TDMA schedule. The TDMA schedule allows nodes to recognize their limitation of this survey is that no taxonomy is carried out to classify
time slots for sending their collected data to CH. these routing protocols. Thus, merely eight variants of LEACH are

4
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

presented. The authors are concentrated on their comparison of a few 3.1.1. Residual energy
factors. However, other main metrics are missing. In CH selection, residual energy is an important metric, which
Mahmuddin et al. in [14] have presented a survey that described impacts on the network performance and the network duration. The
LEACH and some of its variant protocols. This study highlighted the network lifespan will be extended when the selected CH is a sensor
different communication techniques between the CHs and the final des- node with the highest energy among all cluster members.
tination (BS) named: single-hop and multi-hop. The authors mentioned Tong et al. [30] introduced an improved version of LEACH called
the advantages and drawbacks of LEACH and its variant protocols LEACH-B (LEACH-Balanced), where the number of CHs is guarded near-
based on their functionalities and their results. Similar to the pre- optimal. The LEACH-B is a decentralized approach used for cluster
viously mentioned surveys, This survey also has limitations such as formation [31], by enhancing election CHs to make the number of
the review describes ten LEACH-based protocols without offering any CHs optimal given the remaining energy of nodes in the set-up stage.
classification. Thus, authors did not use any metric of WSNs to compare The LEACH-B protocol has three main stages: CH election, clusters
protocols, as residual energy, scalability, energy efficiency, mobility, formation, and data communication.
etc. In the starting, for the first selection of CH, LEACH-B remains
Authors in [15] provided a review that depicts the famous LEACH like the LEACH approach. After that, LEACH-B adds a second phase
protocol and its enhancement protocols. In this survey, protocols are that considers the remaining energy. LEACH-B guarantees a uniform
compared to each other based on the data aggregation, nodes mobility, number of CH in each round, which is calculated through Eq. (2)[30]
and the type of scalability. This review has several disadvantages below, which is estimated between 3% and 5% for the total node
among them is that the authors talked about twelve protocols derived number of 100.
from LEACH without classifying them in taxonomy. Thereby, they 𝑁𝑜.𝑜𝑓 𝐶𝐻𝑠 = 𝑁 ∗ 𝑃 (2)
have not highlighted other factors as well as they did not indicate the
advantages and disadvantages of each discussed protocol. Where P presents the proportion of CHs in the network, and N shows
In [16], Al-Shalabi et al. furnished a survey that presents an the global nodes number in the entire network.
overview of LEACH and its variants. They provided also an overview of The nodes cognize just its own position and the location of the BS
some existing reviews then they introduce their limitations. This survey but do not know the position of other nodes [32]. If CHs number in
put the light on CH election and clusters formation (or authors group the network is inferior to N*P, then a normal node with enough energy
LEACH-based routing protocols into two groups: CHs selection and will be a CH. On the other hand, some low energy CHs will be removed.
clusters formation). Therefore, authors have compared some LEACH- Unlike the LEACH algorithm, LEACH-B guarantees a uniform number
based routing protocols based on the techniques of clusters formation of CHs [33]. Simulation results of the LEACH-B algorithm in [30]
and CHs selection. Thus, the advantages and weaknesses of these demonstrate that LEACH-B imported an enhancement of LEACH, which
protocols are provided as well as their contributions. Thereby, this extends the network lifespan. A flowchart of the LEACH-B is given
survey has compared these protocols depend upon certain aspects such in [30].
The improved LEACH (I-LEACH) is a clustering routing protocol,
as the residual energy, location of nodes, complexity. However, as
which in terms of residual energy, nodes location, and neighbor number
the previous reviews, this survey remains to be developed since many
it picks out CHs [34]. In the I-LEACH approach considers the optimal
metrics of WSNs did not showcase, also some protocols are not very
cluster number in the network 𝐾𝑜𝑝𝑡 described in Eq. (3) [34].
detailed.
As with several clustering routing protocols, the I-LEACH algorithm
The previous surveys mentioned above concentrated either on CH
has three phases namely, CH selection, cluster formation, and data
selection criteria or cluster formation techniques or only talked over
transmission. In the original LEACH algorithm, the CH selection de-
the main contributions, advantages, and disadvantages of LEACH-based
pends only on the pre-set threshold function without considering other
protocols. On the contrary, this survey presents a detailed description of
factors, what influences on the life span of the network. To solve this
the LEACH approach and an overview of precedent surveys. This review
problem, the I-LEACH approach updates the threshold function intro-
suggests a complete classification of LEACH and its variant protocols,
duced in LEACH by putting the residual energy, number of neighbor
which focal point on criteria used on set-up phase (CHs selection and
nodes, and the distance to the BS factors, as defined in Eq. (3) [34]. In
clusters formation) and criteria used on the data transmission phase.
the CH selection stage, every node produces a number between 0 and
1 randomly. If this number is less than the threshold function in Eq. (3)
3. LEACH descendant protocols [34], then it is elected as CH in the present round.
√ √
In this review, we have proposed a new taxonomy of LEACH descen- 𝑁 𝜖𝑓 𝑠 𝑀
𝐾𝑜𝑝𝑡 = √ ∗ ∗ (3)
dant protocols, which focused on CH selection and data transmission 𝜖 𝑑 2
2𝜋 𝑎𝑚𝑝 𝑡𝑜𝐵𝑆
techniques to classify variant protocols of LEACH. Fig. 6 shows the
Once CHs are picking out, every sensor node joined a cluster de-
taxonomy of LEACH-based routing protocols.
pending upon the received signal from CHs. Nodes which are near to
The above taxonomy provides the classification of protocols based each other collect similar input and have a high risk of the overlap [34].
on CH selection and data transmission. Each class is divided into To attack this issue, in the data transfer stage of the I-LEACH protocol,
categories according to the metrics and techniques used. each CH incorporates its data t=received from its cluster members to
decrease significantly the charge of the transmitting additional data.
3.1. CH selection-based protocols Table 1 introduces the advantages and disadvantages of protocols
based on residual energy to elect CH.
CH selection is one of the important phases for extending the
WSNs lifetime. The election of CH influences the energy-efficiency 3.1.2. Centralization
of the WSNs. All routing protocols based on CH selection have the Centralized formation methods are appropriate for networks in
same principle of picking the most efficient node in the network to which the treatment is central control. At each round, nodes send
be a CH (a leader of a cluster). Each protocol selects CH considering some information to the BS, such as the location information and the
one or more specific parameters. Based on these parameters, these residual energy. Based on this information, the sink (BS) makes use
routing protocols are classified into five categories: (1) residual energy, of a particular technique to pick the CH and distributes all nodes
(2) Centralization, (3) the mobility, (4) energy-efficiency, and (5) the to these clusters. A centralized technique guarantees better cluster
distance. These categories are defined in the next parties. segmentation [35].

5
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

Fig. 6. Taxonomy of derived LEACH protocols.

LEACH-C is a centralized clustering algorithm implemented at the CH belongs. To solve this issue, the LEACH-M algorithm can assure
BS [36], which selects CHs according to the residual energy of nodes, that nodes remain in communication with CH or not through the
then forms clusters in the network. LEACH-C uses the centralized TDMA schedule. At the start of each round, CH justifies the absence
method to choose the best CH from nominee CHs, which can ef- of any node by a message request. If the node cannot answer for two
fectively solve the efficiency problem posed by the classical LEACH successive messages, then it is out of the CH coverage [33].
protocol [37]. In the first phase, sensor nodes forward information In the data transmission phase, can be faced with another issue
about their remaining energy and location to the sink. Based on this that if any node or CH is displaced from its cluster, then the data
information, the sink computes the average energy of nodes. Then, will be lost. To resolve this problem, the LEACH-M approach uses the
it decides which nodes are going to be CHs. Hence, nodes that have handover mechanism, which allows nodes to commutate to new CH if
remaining energy more than the average will be selected as CHs in the it is changed the cluster [31].
present round. On the other hand, the rest becomes a normal node. The Authors in [41] proposed an improvement LEACH-M protocol refer
steady phase is identical to that of LEACH. to LEACH-ME (Mobile Enhanced-LEACH), which serves to ameliorate
The centralization method offers popularly a better distribution and LEACH-M. It picks out nodes with minimum mobility comparatively to
selection of CHs. Although, the number of communication between its neighbors as CHs. Each node includes CH transitions it has made
nodes and the sink decreases the life-cycle of the network. Like LEACH, all through the steady stage whilst transmitting data. Nodes transmit
the LEACH-C algorithm sends data directly to the sink in a single hop several transitions to its CH over the TDMA slot [42]. The main aims of
regardless of the distance, which may reduce CH lifetime [38]. Thus, in LEACH-ME is a guarantee that the mobility of CHs is the most minimal
the LEACH approach, the number of CHs changes from turn to others, relative to other nodes in the network, result that when the clusters
while in the LEACH-C approach the number of CHs is predetermined move and change places, the disturbance due to the movement of CHs
for all turns. will be minimal as much as possible.
Table 1 introduces the advantages and disadvantages of the LEACH-
Table 1 introduces the advantages and disadvantages of protocols
C protocol.
based on the mobility of nodes and CHs.
To simulate the LEACH-mobile and LEACH-C algorithm, algorithms
3.1.3. Mobility
are provided in [43].
Mobility is one of the fundamental features of WSN. Mobility in
wireless sensor networks can be classified taking into account: (1)
physical aspects of mobility, which concerns the mobile element and 3.1.4. Energy-efficiency
the type of movement; (2) architectural aspects, such as the entity that Energy-efficiency presents a principal challenge in routing pro-
manages the mobility process [39]. In the CH selection, mobility of tocols to choose CHs. When CHs are picked out with the highest
nodes (or mobility of the sink) can help to choose the most efficient energy-efficiency, the network lifespan is prolonged. There is a signifi-
node as CH and improve the network lifetime. cant difference between the residual energy and the energy-efficiency.
LEACH-M (Mobile LEACH) [40] comprises the mobility factor at the Therefore, routing protocols based on the residual energy to select
conventional LEACH algorithm. This mobility concerns both CHs and CH, they do not necessarily consider the communication charge, the
cluster member nodes over the set-up phase as well as the steady phase, consumption of energy, etc.
whereas the BS is examined as stationary as in the original LEACH. Several clustering routing protocols are based on the energy-
Nodes are selected as CHs based on the lowest attenuation and the efficiency to select CH, such as VH-LEACH, LEACH-T, TB-LEACH, which
minimum mobility. Then, the elected CHs diffuse their position to all are discussed in the following paragraphs.
nodes in their transmission scale [32]. In [44], the authors proposed an enhanced V-LEACH protocol called
The authors of LEACH-M considers that all nodes are initially ho- VH-LEACH. The principal purpose of this protocol is to change the
mogeneous, and the GPS is used to locate sensor nodes. The threshold election of CH. In the VH-LEACH algorithm, CH is selected depending
function of LEACH is used also in LEACH-M to elect CHs, although on the maximum remaining energy [45].
LEACH-M considers mobility of nodes along the data transmission stage Clusters are established and CHs are chosen. Every CH obtains the
contrary to LEACH. information about the remaining energy of cluster members. Afterward,
The mobility method can face a major problem that mobile cluster the CH picks out a VH (Vice cluster-Head) which has the maximum
members may quit its cluster when it is transferring information to its remaining energy within the cluster [44].

6
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

The VH node stays in sleep mode until the residual energy of the Here, 𝑑𝑏𝑠 parameter denotes the distance between a node and the
main CH reaches a threshold. If the CH residual energy achieves this BS, when the distance from the farthest node in a cluster to the BS is
threshold, the VH gets up to play the CH role and picks out its VH. expressed by 𝑑𝑓 𝑎𝑟 .
Authors in [46] introduced a novel LEACH based clustering routing To extend the network lifetime and the scalability, functions de-
protocol named LEACH-T, which allows fixing the number of time slots scribed in Eqs. (5) and (6) are incorporated and multiplied by the
in TDMA of each CH depending on the cluster member nodes number. probability function. The combination function is given in Eq. (7).
In the cluster formation phase, the LEACH-T takes into account the 𝐸𝑐𝑢𝑟𝑟𝑒𝑛𝑡
𝑆𝐶𝑅𝐹 𝑁 = 𝑤1 ( )
remaining energy of nodes, thus determined an optimal number of CHs 𝐸𝑚𝑎𝑥
{ (7)
using Eq. (4) [46]. 𝑑
𝑃
,𝑛 ∈𝐺
√ √ +𝑤2 [ 𝑏𝑠 ] ∗ [ 1−𝑃 ∗(𝑟 𝑚𝑜𝑑(1∕𝑃 )) ]
𝑁 𝜖𝑓 𝑠 𝑀 𝑑𝑓 𝑎𝑟 0 otherwise
𝐾𝑜𝑝𝑡 = √ ∗ ∗ (4)
𝜖 2
𝑑𝑡𝑜𝐵𝑆
2𝜋 𝑎𝑚𝑝 Here, 𝑤1 +𝑤2 =1, with 𝑤1 is the weight that is given according to
the preference for energy, and 𝑤2 is the weight that is given according
In the steady phase, the T-LEACH involves variable time slots for
to the preference for distance.
distinct CHs [46]. Instead of a fixed allocation of time slots to each CH,
EADCR (Energy Aware Distance-based Cluster Head selection and
the LEACH-T algorithm uses a dynamic technique to allocate time slots
Routing) is proposed by [49] to prolong the network lifetime of WSNs
in the TDMA schedule according to the number of cluster members. based on the FCM (Fuzzy C-Means), the residual energy of nodes, the
That means, if the quantity of nodes is little, then only a few time slots Euclidean distance from the BS, and cluster centroid. In the EADCR,
are allotted to the CH, while if the cluster member nodes is bigger, so clusters are composed through making use of the FCM approach at
giant time slots should be assigned [13]. Thereby, to transfer data from the BS. Then, each cluster selects its CH using a fitness function. This
CHs to the sink, it uses either single-hop or multi-hop. function depends upon the Euclidean distance between the node and
TB-LEACH (Time-based Cluster-Head Selection Algorithm for the BS, the distance from the node to its CH, the initial energy, and the
LEACH) is defined in [47], which modifies the CH selection of the residual energy of nodes, as defined in Eq. (8) [49].
LEACH approach to maximize the life-cycle of the network. TB-LEACH ⎧𝛼𝑥1 + (1 − 𝛼)𝑥2 , (𝐸𝑟𝑒𝑠 >= 𝐸𝑡ℎ ) & (𝑑𝑛2𝐵𝑆 > 𝑑𝑛2𝐶 )
includes two main phases: Set-up and steady stages. In the first phase, ⎪
𝐹𝐶𝐻 = ⎨𝑥1 , (𝐸𝑟𝑒𝑠 < 𝐸𝑡ℎ ) ∥ (𝑑𝑛2𝐵𝑆 = 𝑑𝑛2𝐶 ) (8)
instead of the random selection of CHs in LEACH, the TB-LEACH
⎪ ′
algorithm selects CHs to depend upon a random timer. Unlike LEACH, ⎩𝛼𝑥1 + (1 − 𝛼)𝑥2 , (𝐸𝑟𝑒𝑠 >= 𝐸𝑡ℎ ) & (𝑑𝑛2𝐵𝑆 < 𝑑𝑛2𝐶 )
the number of CHs in each turn is fixed and represented by a counter. Here, 𝐸𝑟𝑒𝑠 represents the residual energy of each node, while 𝐸𝑡ℎ
At the start of each epoch, a random timer is generated by every node. expresses the threshold energy described in Eq. (9). The Euclidean
If the random timer is expired and the counter does not achieve the distance from the node n to the BS; and its centroid are presented
number of CHs, the node diffuses a CH-ADV advertisement message to by 𝑑𝑛2𝐵𝑆 , 𝑑𝑛2𝐶 , respectively. 𝑥1 represents the ratio of the remaining
announce at all nodes that it becomes a CH in the current turn using energy to the commencing energy of the sensor node. 𝑥2 and 𝑥′2 are
CSMA [19]. Even if, the node will be a non-CH node (normal node). expressed in Eq. (10). Finally, the authors are chosen the value of 𝛼
After that CHs are determined, the rest of the tasks are similar to the equal to 0.5.
LEACH algorithm. 𝐸𝑡ℎ = 𝑙(𝐸𝑒𝑙𝑒𝑐 + 𝐸𝐷𝐴 )𝑛𝑐 (9)
Generally, The TB-LEACH algorithm improves LEACH protocol by
fixing the CHs number and selecting CHs depending on a random timer. Where, l introduces the number of bits, 𝐸𝑒𝑙𝑒𝑐 describes the energy
Simulation results in [47] showed that the TB-LEACH enhances the consumed in transmitting or receiving 1-bit data. The energy consump-
network life span by up to 30% compared to the conventional LEACH tion during the data aggregation process is expressed by 𝐸𝐷𝐴 . Finally,
the 𝑛𝑐 parameter is the number of cluster members.
protocol.
Table 1 introduces the advantages and disadvantages of protocols 𝑥2 = |(𝑑𝑛2𝐵𝑆 − 𝑑𝑛2𝐶 )∕𝑑𝑛2𝐵𝑆 |,
(10)
base on energy-efficiency to improve the set-up phase. 𝑥′2 = |(𝑑𝑛2𝐶 − 𝑑𝑛2𝐵𝑆 )∕𝑑𝑛2𝐶 |
The CHs selection is simultaneously from cluster members. Each
3.1.5. Distance cluster elects the node which has the highest fitness value as a CH.
Distance between CH and the BS also has a vital impact on energy The authors have implemented the rotational-based clustering concept
consumption. When the CH is very far from the BS, it results in a used in the LEACH protocol. In the EADCR protocol, according to the
major issue in energy consumption. The CH sends aggregated data to residual energy and the Euclidean distance, a new approach for packet
the BS, so the distance from the CH to the BS will influence the data routing is proposed.
transmission process. Therefore, the distance factor is very important BN-LEACH (Bayesian Network -LEACH) [50], the authors used the
in selecting CH. model of Bayesian Network (BN) to pick out CHs in the network. In this
Numerous clustering routing protocols are based on the distance to approach, three main factors are considered, the density of nodes, the
pick out CHs, such as Improved-LEACH, EADCR, and BN-LEACH. distance to the BS, and the residual energy. The BN-LEACH operations
are relatively the same as the conventional LEACH approach, except
Improved-LEACH is one of the clustering routing protocols based on
that the CH selection reposes on the BN model. This model computes
the distance to select CHs in the network. Authors in [48] proposed
the probability to become CH for every node. The BN-LEACH routing
this modified LEACH protocol by considering two parameters when
protocol serves to prolong the network lifetime by extending the first
selecting CH: the residual energy of nodes and the distance to the BS.
node death and distributing the energy load among the sensor nodes. To
Authors use Eqs. (5) and (6) [48] to calculate the residual energy distribute CHs uniformly in the network, the convenient CHs are chosen
of each node and the distance from CH to the BS, respectively. under a dynamic zoning technique and a greedy method to distribute
𝐸𝑐𝑢𝑟𝑟𝑒𝑛𝑡 cluster-heads uniformly. The area is divided into three regions, far,
𝑆𝐶𝑅𝐹 𝑁1 = (5) central, and near. The central area has the highest prior probability,
𝐸𝑚𝑎𝑥
near having medium and far has lowest [51].
Here, 𝐸𝑚𝑎𝑥 presents the initial energy of the node, while 𝐸𝑐𝑢𝑟𝑟𝑒𝑛𝑡 The discussed improved LEACH hierarchical routing protocols based
expresses the residual energy of each node. on the CH selection to ameliorate the LEACH approach are summarized
𝑑𝑏𝑠 in Table 2 according to the research domain, network type, modeling
𝑆𝐶𝑅𝐹 𝑁2 = (6) parameters of each protocol, and the complexity of the algorithm.
𝑑𝑓 𝑎𝑟

7
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

Table 1
Advantages and disadvantages of LEACH-based protocols (class 1).
Protocol Strengths Limitations
Improves the life-cycle of the network through using the TDMA Assumes non-uniform deployment of CH nodes.
schedule.
LEACH
Equilibrates the energy consumption. Random election of CH.
Decreases the no. of transmission packets. The quantity of nodes in every cluster is not evenly distributed.
Diminishes energy dissipation of nodes. Uses the single hop.

Uniform No. of CHs.

LEACH-B Considers remaining energy after the first turn. Growth in overhead cost.
Enhances the network lifespan.

I-LEACH Elects CH based on the remaining energy and nodes location. CH integrates collected data to reduce the cost of supplementary
data transmission, which is not practice for nodes that
receive different data.

Reaches a high no. of turns in a petty network zone. Needs location information of nodes.
CH selection according to residual energy. Centralization introduces an overhead on the sink.
LEACH-C
The centralization method offers a better CHs distribution. Transmission using single hop produces an extra overhead.
Introduces an optimal no. of clusters.

Mobility of both CHs and non-CH node.

LEACH-M Supplementary overhead.
A very high energy-efficiency.

Supports the mobility of nodes.

LEACH-ME Increased overhead.
Selects CHs depend upon the minimum mobility.

CHs election on the basis of the remaining energy. Supplementary treatment for VH node.
VH-LEACH
Replaces CH by VH node when the main CH achieves a threshold. Uses the single hop to direct data from CHs to the BS.

Fixes the no. of time slots in TDMA based on the no. cluster CHs are selected randomly.
LEACH-T
members.
Offers an optimal no. of clusters. Data transmission from CHs to the sink using a single hop.

Increases the life-cycle of the network. CHs selection based on a random timer.
TB-LEACH
The no. of CHs is fixed.

Improved-LEACH Amelioration of the network performance by a well selection of CH. Complexity of calculation.

Table 2
General characteristics of LEACH-based protocols (Class 1).
Protocol Year Research domain Network type Modeling Parameters Complexity
LEACH 2000 Load balancing homogeneous Energy saving Low
LEACH-B 2010 balancing the number of CHs homogeneous Remaining energy Yes
LEACH-C 2002 Cluster head changing homogeneous Remaining energy and nodes position from the sink Yes
LEACH-M 2006 Mobility of nodes homogeneous Mobility factor Yes
LEACH-ME 2008 Mobility and status of nodes homogeneous Mobility factor Yes
I-LEACH 2013 Energy-Efficient homogeneous Remaining energy, Neighbor number and nodesś location No
VH-LEACH 2015 Multiple CHs in the cluster homogeneous Remaining energy Yes
LEACH-T 2014 Optimal Number of CHs homogeneous Remaining energy Yes
TB-LEACH 2008 Cluster head changing homogeneous Random timer Yes
Improved-LEACH 2016 Energy-efficiency homogeneous Distance from node to BS Yes
EADCR 2020 Energy-efficiency homogeneous Euclidean distance and Residual energy Yes
BN-LEACH 2014 Load balancing homogeneous Density, distance and residual energy No

3.2. Data transmission-based protocols Once clusters are established, they are fixed and there is no re-
clustering stage again at the starting of each epoch [53]. Therefore,
Data transmission is the second significant phase for prolonging clusters are fixed and just turnaround the role of CH in their cluster.
the WSNs lifetime. This phase represents data communication between Furthermore, the steady phase stays unchangeable compared to the
normal nodes and their CH, and between CHs and the BS. So, it largely LEACH approach. LEACH-F reassures the reduction of the set-up phase
affects the energy consumption in the network. The principal idea in by fixing the clusters at the beginning of the phase. However, LEACH-F
all data transmission-based protocols is to route data either in a single has many disadvantages, among them:
hop or in multi-hop or both techniques. Consequently, these techniques
classified these routing protocols into three categories, called, single- • Fixing clusters may oblige nodes to remain in a cluster even if the
hop, multi-hop, both single and multi-hop, which discussed in the next CH is very further.
parties. • No flexibility to remove or add nodes to the network after the
clusters are provided.
3.2.1. Single hop • Nodes cannot regulate their conduct at the death of nodes.
In this category, protocols are based on the single hop technique to • The mobility of nodes cannot be managed.
route data from CHs to the BS.
Authors in [52] introduced the LEACH-F (Fixed number of cluster- The CH expends more energy than other nodes due to its tasks.
LEACH) protocol, which forms clusters and selects CH using a central- To save this energy consumption, authors in [54] introduced a new
ized approach like the LEACH-C algorithm. mechanism called V-LEACH (Vice Cluster-Head-LEACH to avoid the

8
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

Fig. 8. LEACH-Cell architecture.

its cluster [58]. Following the number of neighbors, each node can
Fig. 7. V-LEACH architecture. define its status as being an active or stay in a sleep mode over the
current turn. Hence, sensor nodes that have low densities remain alive
as long as possible [57].
quick death of CHs. This protocol designed a new scheme as shown W-LEACH decentralized uses the density of sensor nodes to decide
in Fig. 7. the node status either active or sleep status. Thus, in every cluster it
Once clusters are formed, the V-LEACH algorithm elects a node allows just a part of nodes to forward data to the CH in the epoch.
within the cluster as a vice-CH depending on the remaining energy. Consequently, W-LEACH decentralized algorithm can decrease the re-
In the beginning, normal nodes transfer their collected data to the CH, dundant data CHs and augment the network lifetime. However, it has
then this latter to the sink. When the CH dies, the vice-CH takes the some disadvantages, such as it can work just only in-network where the
role of the CH. This protocol increases the efficiency of the CH election collected data is the same.
by selecting a node as a vice-CH, which replaces the CH in the event of A WD-LEACH diagram is provided in [57].
death. Thus, it assures that data is always transmitted to the sink even Table 3 introduces the advantages and disadvantages of protocols
if a CH dies in a specific round [55]. base on a single-hop technique to route data from CHs to the BS.
An improved centralized algorithm based on LEACH named W-
LEACH (Weighted-LEACH) is proposed in [56], where the BS is the 3.2.2. Multi hop
unique component that responsible for picking active sensor nodes The multi-hop technique reduces the energy cost used by all nodes
and sleeping sensor nodes. This centralized data aggregation protocol along the route. In this category, protocols are based on the multi-hop
includes two phases like LEACH: Set-up and steady phase. In the first technique to route data from CHs to the BS.
phase, authors change the P definition to become the percentage of the Among the protocols used to improve network coverage in sensor
maximum quantity of CHs in place of the current number of CHs as it networks are LEACH-Cell [59], which divides the network into sections
is set in the classical LEACH protocol. named cells. Every seven neighboring cells formed a cluster with a CH.
At the beginning of each round, W-LEACH calculates a weight value Each cell contains some sensor nodes and a cell-head that is elected
Wi that it is attributed to every sensor node, without considering if this within the cell to be a head as we can see in Fig. 8.
node was a CH or not in the precedent turns [55]. Residual energy and The cell-head has the same function as CH in the LEACH approach.
density of nodes are two principal elements, which calculate the weight That means, it creates a TDMA schedule, allocates a time slot to each
of the node (Wi) [13]. Wi value is calculated applying Eq. (11) [56]. cell member, aggregates data from cell members through the TDMA
{
𝑒𝑖 ∗ 𝑑𝑖 , 𝑖𝑓 𝑑𝑖 > 𝑑𝑡ℎ𝑟𝑒𝑠ℎ schedule, then forwards the received data to the CH belongs. The CH
𝑇 (𝑖) = (11) repeats the same procedure as cell-head in order to gather data from its
𝑑𝑖 otherwise
cell-heads, then it propagates it to the sink through the optimal path.
Where the density of each node is 𝑑𝑖 No re-celling and re-clustering are carried out other times because they
1 + 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑎𝑙𝑖𝑣𝑒 𝑛𝑜𝑑𝑒𝑠 𝑖𝑛 𝑟𝑎𝑛𝑔𝑒 𝑟 keep the same in all along the life-cycle of the network once formed,
𝑑𝑖 = (12) only CHs and cell-heads switch dynamically [59].
𝑛
Where r is the range that is attainable by sensor node i. 𝑒𝑖 presents The principal purpose of LEACH-Cell is to switch off the nodes trans-
the residual energy of node i, and 𝑑𝑡ℎ𝑟𝑒𝑠ℎ is a density threshold to define mitter when the cell-head sends data to the CH. Thus, the generation
the set of sensors in very low-density areas [56]. of cell-heads decreases communication cost between nodes and the
CHs are chosen depend upon the value of Wi, such that nodes with CH [60].
high Wi have a high chance to be selected as CH. The pseudo-code for selecting a new cell-head and a new cluster-
Once CHs are formed, clusters are established such that every sensor head is shown in [59].
node is attributed to its closest CH. Owing to the random selection of CHs in the LEACH protocol, some
Hnini et al. in [57] improved W-LEACH algorithm by W-LEACH CHs may be gathered on a specific network surface. Consequently, some
decentralized (WD-LEACH). The WD-LEACH is a decentralized proto- sensor nodes joined a CH, nonetheless many sensor nodes cannot be
col, which serves to prolong the network lifetime in WSNs by using connected to any cluster. These sensor nodes are examined as orphan
the density of the nodes to split the network into two groups: active nodes. Hence, orphan nodes’ data cannot achieve the BS. To solve this
nodes and sleeping nodes. Tasks in W-LEACH decentralized is done into situation, authors in [61] developed O-LEACH (Orphan-LEACH), which
epochs in two stages as well as in W-LEACH and LEACH protocols. allows these nodes to transmit their data to the sink.
Every epoch starts with the set-up stage. In this stage, CHs are O-LEACH addressed two scenarios, the first one considers a cluster
picked out and clusters are established. Afterward, the permanent stage member as a gateway, which is capable to take on the role of CH of
is coming, that characterized by data transmission. In this phase, every orphan nodes. Then, a number of orphan nodes are connected to the
non-CH node can precise its neighbor and the distance corresponding to gateway, and this latter forwards aggregated data to its CH belongs as
every neighbor depending on the density of nodes. Then, it constructs shown in Fig. 9.

9
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

then lead information to the BS through these groups [68]. But, the
MHT-LEACH cannot be effective for the large-scale networks [69].
Instructions of the algorithm and a flowchart of the MHT-LEACH are
given in [64].
To improve MHT-LEACH, authors in [66] developed a new exten-
sion of MHT-LEACH which is referred to as IMHT-LEACH (Improved
Multi-Hop Technique-LEACH), that distributes CHs into multi-levels
instead of using two levels in MHT-LEACH to classified all CHs in the
network [67,70].
Fig. 9. Scenario 1 general and more precise of orphan nodes in the network. Each level has a diameter of 𝑑0 ∕2 from the BS. e.g.:

• Level 1 : distance to the BS (𝑑𝑡𝑜𝐵𝑆 ) < 𝑑0 ∕2.

• Level 2: 𝑑0 ∕2 <= (𝑑𝑡𝑜𝐵𝑆 ) < 𝑑0 .
• Level 3: 𝑑0 <= (𝑑𝑡𝑜𝐵𝑆 ) < 3𝑑0 ∕2.
• ... Until level n.

As MHT-LEACH, in IMHT-LEACH, each CH has its own routing table

which it transfers its data based on it [71]. Data is routed between CHs
from higher levels to lower ones until it achieves the sink [67].
IMHT-LEACH divides the set of CHs into multi-levels in the network
depending on the distance from the BS to obtain balanced energy
Fig. 10. Scenario 2 of orphan nodes in the network. throughout the network and extend the network life cycle [68]. The
simulation results show that IMHT-LEACH avoided MHT-LEACH disad-
vantages. Consequently, the IMHT-LEACH technique is more adapted
The second scenario has been proposed to form a private orphan for a wide network environment [72]. Instructions of the algorithm and
nodes cluster with a special CH [16]. This scenario is applied when the a flowchart of the IMHT-LEACH are given in [66].
quantity of orphan nodes overtakes the number of cluster participants Alnawafa et al. proposed a dynamic technique named DMHT-
of CH. The private cluster consists of a number of orphan nodes and a LEACH (Dynamic Multi-Hop Technique-LEACH), considering the CH
CH’ (special CH). The CH’ is the first orphan node that has reached the residual energy over the turn [73], to enhance the IMHT-LEACH
gateway. Therefore, the CH takes its role as a CH, then it broadcasts protocol, which uses a static technique to route data from CHs to the BS.
information to its cluster members. Afterward, it regroups data, then Hence, CH in the DMHT-LEACH can change its data route dynamically
sends it to the CH through a gateway as shown in Fig. 10. [61]. many times in the same epoch [74]. The DMHT-LEACH has four stages
to do its job: the initial stage, the announcement, the selection, and the
In summary, in both scenarios, data of orphan nodes reach the BS.
routing stage.
The multi-hop method between nodes and the BS improves the
The first stage has been done as the set-up stage in the LEACH
distance of communication. Authors in [62] have proposed the MH-
approach, which serves to elect CHs and establish clusters. Then, every
LEACH (Multi-Hop-LEACH) to pick out the optimal target to transfer
CH generates its own TDMA schedule. This schedule contains allocation
data from CH toward the BS using intermediate CHs. The principal goal
for cluster member nodes and allocations for CHS neighboring. Once
of this protocol is the communication between CHs and using multi-hop the first phase was done, CHs are classified into levels according to
to reach the sink. their distance toward the BS. The wide of each level is 𝑑0 ∕2.
The MH-LEACH is established in two phases (the initial phase and The second stage is characterized by an announcement message that
the steady phase). CHs and clusters are determined in the same way it is broadcast by CHs. This message contains information about the CH
as in LEACH. Afterward, CHs broadcast announcement messages to the ID, the CH coordinates, the distance between CH and the BS, then the
network. Each CH creates its routing table according to the received CH current energy [71].
signal power. In the third stage, each CH establishes its routing table based on
After routing table generation, CHs forward their tables to the BS to information collected from lower level CHs in the second stage.
check whether CH can be in the path of another. Then, the BS returns When routing tables are formed, in the routing stage, the DMHT-
to pack the checked tables to the CHs. [62] When a data packets did LEACH algorithm arranges these routing tables and each CH chooses
not accept by any CH, the source CH essays another route in its routing its optimal route to the BS depending on the residual energy of its
table [12]. To simulate the MH-LEACH see algorithm in [63]. neighbors and their distances to the BS. Therefore, a cost function is
The MHT-LEACH (Multi-Hop Technique-LEACH) [64] is a protocol calculated Eq. (13) [71] to organize routes in the routing table.
based on the LEACH approach which discusses routing by split the
𝐶𝑁𝐶𝐻 (𝑖) = 𝐶𝑁𝐶𝐻𝑃 (𝑖) + 𝛼.𝐶𝑁𝐶𝐻𝐸 (𝑖) (13)
network into two levels depending on the path length between CHs
and the BS. Where
In MHT-LEACH, the CH selection and the cluster formation are 𝐷𝑖2
similar to the LEACH set-up phase. Instead of forwarding CHs data 𝐶𝑁𝐶𝐻𝑃 (𝑖) = (14)
𝑆𝐶𝐻𝐷−𝐵𝑆 − 𝑁𝐶𝐻𝑖𝐷−𝐵𝑆
immediately to the sink [65], MHT-LEACH divided CHs into two groups
Where, 𝐶𝑁𝐶𝐻𝑃 represents the distance cost for the NCH (Next CH),
(internal and external) according to the location betwixt CHs and the
𝐶𝑁𝐶𝐻𝐸 refers to the energy cost for the NCH. 𝐷𝑖 introduces the distance
sink. If this distance is strictly lower than 𝑑0 , this CH belongs to the from the SCH (Source CH) to the NCH. 𝑆𝐶𝐻𝐷−𝐵𝑆 refers to the distance
internal group which communicates directly with the sink. On the other between the SCH and the BS, while 𝑁𝐶𝐻𝑖𝐷−𝐵𝑆 indicates the distance
side, if the CH is located equal to or upper 𝑑0 , the CH belongs to the of the NCH to the sink. The parameter i presents the member in the
external group [66]. group of neighboring CHs which are selected by the SCH in this turn.
Every external CH is able to generate a routing table based on the Finally, the 𝛼 criterion shows the relative weight.
announcement messages received from the internal CHs [67]. Then, The advantage of the DMHT-LEACH is that it uses a dynamic
each external CH picks the lowest route of its routing table in order technique to route data, so the route in the routing table can be changed
to transmit its data either directly to the BS or to another CH from frequently when the cost function change.
the internal group. The MHT-LEACH protocol serves to balance energy Table 3 introduces the advantages and disadvantages of protocols
charge in the network by distributing the set of CHs into two groups, base on the multi-hop technique to route data from CHs to the BS.

10
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

Fig. 12. LEACH-TL architecture.

sink [12]. The LEACH-TL is based on the LEACH approach to elect CHs
and form clusters. Afterward, if the remaining energy of the CH is less
than the average (see Eq. (15)) [26] or its distance to the sink is more
Fig. 11. A-LEACH architecture.
than the average (see Eq. (16) [26]), then this CH requires a secondary
CH within its cluster. This latter will be with maximum residual energy.
Furthermore, if the CH remaining energy is more than the average, so
3.2.3. Single and multi hop this cluster does not require a secondary CH [25].
In this category, protocols are based on both the single hop and the
multi-hop techniques to route data from CHs to the BS. ∑
𝑁
𝐸𝑎𝑣𝑒 = 𝐸𝑐𝑢𝑟 (𝑖) (15)
Advanced-LEACH (A-LEACH) [75] is proposed to solve a drawback
𝑖=1
of LEACH protocol that the node which is selecting to be CH for the ∑
𝑁
current turn consumes much energy than other normal nodes. A-LEACH 𝑑𝑎𝑣𝑒 = 𝑑𝑖 (16)
is a heterogeneous energy algorithm. Its main aims are prolonging the 𝑖=1
network lifetime or extending the stability epoch of the network by Here, 𝐸𝑐𝑢𝑟(𝑖) introduces the current energy of the CH, and 𝐸𝑎𝑣𝑒
reducing the risk of nodes failure by using the heterogeneity attribute presents the average energy. The 𝑑𝑖 refers to the distance between the
parameters [75]. A-LEACH executes its tasks in turns like the LEACH CH and the BS, and 𝑑𝑎𝑣𝑒 presents the average distance.
algorithm and every turn is split into two stages. A synchronized clock Fig. 12 presents the architecture of the LEACH-TL protocol.
is used to indicate to nodes the beginning of each turn. To simulate the LEACH-TL approach, an algorithm is given in [43].
In A-LEACH, there are two types of nodes: CAG are nodes that have L-LEACH (Energy Balanced) [76] is an energy multi-hop routing
more energy than others, which will be chosen as CHs or gateways, and protocol, which takes into account only the distance. Unlike LEACH,
the rest of the network nodes represent normal nodes. LEACH-L is adapted for wide range WSNs and the optimal number of
In the set-up stage, based on the intended probability of CHs, every the hop is taken.
node determines its probability of becoming a CH. In the steady stage,
CHs communicate to the sink in various ways: single-hop and multi-
when the collected data achieves the CH, this latter regrouped it, then
hop according to the distance. They communicate directly to the BS if
forwarded it to the BS using one of the two ways: either it chooses the
they are closed to it. On the other hand, when they are further to it they
nearest gateway (CAG node), which will direct the data to the sink if
communicate through multi-hop techniques, and the shortest distance
the CH is normal or it routes the data directly to the sink if it is a CAG
of communication is restricted.
node. as we can show in Fig. 11.
The LEACH-L consists of two stages like LEACH. At every turn, a
The benefits of A-LEACH are:
novel CHs are picked out, and the charge is diffused and well-regulated
• It is a decentralized algorithm where clusters are configured betwixt the set of network nodes. LEACH-L covers up a giant region
independently of the sink. than Mobile-LEACH due to the fact LEACH-L allocates equal energy
• It reduced the information quantity that is sensed to the sink amongst all the sensor nodes in the pre-period [31].
through the data fusion. Table 3 introduces the advantages and disadvantages of protocols
• Even if all normal nodes die, the CAG nodes keep on forwarding base on single-hop and multi-hop techniques to direct data from CHs
information to the BS. to the BS.
• The gateway can decrease the energy consumption and the chance The discussed enhanced LEACH hierarchical routing protocols based
of nodes failure, then it prolongs the CHs lifetime. on data transmission techniques to improve the LEACH algorithm are
summarized in Table 4 according to the research domain, network
LEACH-TL [26] solves the unequal energy distribution issue posed
type, modeling parameters of each protocol, and the complexity of the
by the random selection of CHs in the LEACH protocol. The principal
algorithm.
purpose of LEACH-TL is to balance the energy consumption in the
network.
Due to the aleatory election of CH, after the first turn, CHs had 3.3. Protocols based on CH selection and data transmission
different remaining energy. So, if a node with low energy or so far
from the sink was elected as CH, this latter dies rapidly than other In clustering routing protocols in WSNs, both CH selection and
nodes. In this case, LEACH-TL solves this issue by selecting a secondary data transmission are important and have a significant influence to
CH in the cluster that receives data from cluster members, regroups increase the network life-cycle. Protocols based on CH selection and
them, then sends them to the primary CH, while the primary CH is data transmission consider the multi-hop technique and some specific
responsible to forward data received from the secondary CH to the metrics to select CH, such as residual energy.

11
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

Table 3
Advantages and disadvantages of LEACH-based protocols (class 2).
Protocol Strengths Limitations
Offers better CHs distribution using centralization. Clusters are fixed at the beginning.
LEACH-F Reduces the delay of set-up stage. No re-clustering stage.
Once clusters are established, adding or taking off nodes is impossible.

Replaces CH by Vice-CH in case of death of the main CH. The early death of the CH when it is further from the BS.
V-LEACH Extend the lifespan. Additional treatment for vice CH.
There is no require to elect a novel CH each time when CH dies.

W-LEACH Economy energy consumption. Random CH selection.

Dividing the network into active and sleep nodes

WD-LEACH
Allows only a part of sensor nodes to communicate with the CH. Will not be effective in networks where the data collected is different.

Good coverage of the network. Random selection of CHs and cell-heads.

LEACH-Cell
Energy efficiency increase. Various type of head (CHs and cell-heads) provides overload of
control packet.

Orphan nodes can send their data to the BS. Random selection of CHs.
O-LEACH
Good coverage of the network. Communication with the BS in a single hop.

A very high energy-efficiency. Random CHs election.

MH-LEACH
Use the multi-hop method. CHs that group data of other CHs will consume more energy.

A very high energy-efficiency. Selects CHs without taking into account the energy.
Using the multi-hop technique between CHs. The no. of CHs in each turn may be not the same.
MHT-LEACH
Classification of CHs into two groups based on the distance to the sink. Only two levels, CHs at the edge of the network
consume more energy.
The No. of nodes in clusters is not equally.

Uses the multi-hop technique between CHs. Random election of CHs.

IMHT-LEACH
Splits the CHs into multi levels. Considers only the distance to the BS.

Divides network into multi levels depend upon the distance to the BS. Random selection of CHs.
DMHT-LEACH Each CH generates its routing table on basis of the distance and The no. of CHs in every epoch may be not the same.
the remaining energy.
The route will change once the cost function change. The no. of nodes in clusters is not equal.

Extend the stability period. Random CHs election.

A-LEACH Reduces the risk of nodes failure. Based on the single hop for data transmission.
Additional treatment of CAG nodes.

Energy harvesting. Growth in overhead cost.

LEACH-TL
Multiple CHs.
Considers the remaining energy and distance for selecting Single hop to send data from CHs to the BS.
the secondary CH.

Is adapted for large networks. It takes into account only the distance.
L-LEACH
Uses an optimal number of hop. CHs with less residual energy can die quickly.
Regulated network load.

Table 4
General characteristics of LEACH-based protocols (Class 2).
Protocol Year Research domain Network type Modeling Parameters Complexity
LEACH 2000 Load balancing homogeneous Energy saving Low
A-LEACH 2010 Energy saving heterogeneous Energy harvesting ability Yes
LEACH-Cell 2012 Division of the network homogeneous Remaining energy and shortest route Yes
LEACH-F 2013 Fixing the number of clusters homogeneous No. of clusters Yes
LEACH-TL 2013 Energy saving and Multiple CHs homogeneous Remaining energy and distance to the sink Yes
V-LEACH 2013 Multiple CHs in the cluster homogeneous Remaining energy Yes
O-LEACH 2016 Division of the network homogeneous Remaining energy and distance from a orphan node to the gateway Yes
W-LEACH 2013 Division of the network homogeneous Weight value Yes
WD-LEACH 2014 Division of the network homogeneous Density of nodes Yes
L-LEACH 2013 Network Coverage homogeneous Distance Yes
MH-LEACH 2014 Multi-hop technique homogeneous Distance from CHs to the sink Yes
MHT-LEACH 2016 Division of the network and multi-hop homogeneous Distance from CHs to the sink and consumed energy Yes
IMHT-LEACH 2017 Multi-level and multi-hop homogeneous Distance and consumed energy Yes
DMHT-LEACH 2017 Multi-level and dynamic route homogeneous Residual energy and distance Yes

E-LEACH (Enhanced-LEACH) [77] is an enhanced version based on At the set-up level, the E-LEACH accounts that the remaining energy
LEACH. This protocol made changes to the set-up phase and the steady is the principal metric [53]. Therefore, from the second round, it elects
phase. CH according to the residual energy of nodes. That means in each

12
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

Fig. 13. E-LEACH architecture.

round, the node has more remaining energy will be candidate as CH

Fig. 14. BRE-LEACH architecture.
while nodes with few residual energy will be as normal nodes [55].
Authors are defined as a new threshold function [77] as follow:
{ 𝑃ℎ𝑒𝑎𝑑 𝐸
1−𝑃ℎ𝑒𝑎𝑑 ∗(𝑟 𝑚𝑜𝑑(1∕𝑃ℎ𝑒𝑎𝑑 ))
∗ 𝐸𝑐𝑢𝑟𝑟 , 𝑛 ∈ 𝐺 then forwards it to the sink. On the other hand, the non-root-CH use
𝑇𝑛𝑒𝑤 (𝑛) = 𝑖𝑛𝑖𝑡 (17)
0 otherwise the multi-hop metric to achieve the parent CH.
Fig. 14 illustrates the architecture of the BRE-LEACH algorithm.
Where Simulation results in [25] demonstrated that the BRE-LEACH algo-
√ √
𝐸𝑓 𝑠 rithm is better than LEACH in decreasing the cost of energy consump-
𝑁 𝑀
𝑃ℎ𝑒𝑎𝑑 = (18) tion and prolonging the network lifespan by 55.73% compared to the
2𝜋 𝐸𝑚𝑝 𝑑 2 ∗𝑁
𝑡𝑜𝐵𝑆 LEACH algorithm.
At the steady level, to improve data communication directly to the To improve the random selection of CH in DMHT-LEACH which in-
BS in a single hop, the E-LEACH protocol proposed to pick out a CH fluences the network life cycle. Authors in [74] developed an improved
with maximum residual energy as a root-CH which aggregates data version of DMHT-LEACH called EDMHT-LEACH (Enhanced Dynamic
from other CHs, then sends them to the BS. Besides, the other CHs use Multi-Hop Technique-LEACH), which characterized by a new method to
the optimal path betwixt CHs to reach the root. elect CH and forming clusters. Thus, this new approach provides some
Fig. 13 presents the architecture of the E-LEACH protocol. modifications in the dynamic technique to choose routes. Like IMHT
Authors in [25] introduced an enhanced LEACH protocol, namely and DMHT, the EDMHT protocol divided the network into multi-levels
BRE-LEACH (Balanced Residual Energy-LEACH), which improved the around the sink. The wide of each level is 𝑑0 ∕2. Similar to DMHT, this
network life cycle, the stability period, and decreased the cost of energy new version has four stages. The changes made by the EDMHT-LEACH
consumption. The CH consumes more energy than other network nodes technique are as follows:
because it gathers all cluster member’s data. Therefore, the random se-
• CH selection according to the residual energy of node applying
lection of CHs in the LEACH approach without considering the residual
the following equation (20)[74].
energy of nodes causes an energy imbalance, and some nodes can finish { 𝑃 𝐸
their energy quickly than other nodes. So, this problem is solved by the 𝑚𝑎𝑥( 1−𝑃 ∗(𝑟 𝑚𝑜𝑑(1∕𝑃 ))
∗ 𝑟𝑒𝑠𝑖𝑑𝑢𝑎𝑙
𝐸𝑚𝑎𝑥
, 𝑇𝑚𝑖𝑛 ), 𝑛 ∈ 𝐺
BRE-LEACH algorithm through electing CHs according to the remaining 𝑇 (𝑖) = (20)
0 otherwise
energy and use the threshold introduced in Eq. (19) [25].
{ 𝑃 𝐸 • The number of nodes in each cluster (𝑁0 ): EDMHT assumes that
1−𝑃 ∗(𝑟 𝑚𝑜𝑑(1∕𝑃 ))
∗ 𝐸𝑟𝑒𝑠 , 𝑛 ∈ 𝐺
𝑇1 (𝑛) = 0 (19) the number of nodes within the cluster is limited to (𝑁0 ) defined
0 else in Eq. (21) [74] and cannot exceed it [74].
Where the chance of the number of CH in the entire network is 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑠𝑒𝑛𝑠𝑜𝑟 𝑛𝑜𝑑𝑒𝑠
𝑁0 = (21)
presented by P, r is the present turn, 𝐸𝑟𝑒𝑠 introduces the residual energy 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝐶𝐻𝑠
of nodes while 𝐸0 defines the beginning energy, then G shows the
• The appearance of IN (Independent Nodes) in the network: these
collection of nodes which have not been CH yet in the last 1/P turns.
nodes represent nodes that have not joined any CH in the cluster
As the above-discussed protocols, BRE-LEACH works in turns, and
formation phase.
each turn is split into four phases namely: Cluster set-up, TDMA
scheduling, Root CH selection, and Data transmission phase. Once CHs are elected, clusters are established and each node recog-
Once CHs are elected depending on the remaining energy, each nized its role in the network, CHs and INs create their routing table
sensor node joined its cluster depending on the received signal from depending on the announcement messages. Then, routes in routing
CHs. Thus, every CH provides a TDMA schedule for its cluster mem- tables are organized based on the cost function. Hence, as in DMHT, the
bers. Afterward, the BRE-LEACH algorithm chooses the CH which has lowest cost routes will be used to direct CH or IN data to the BS. Then,
residual energy more than the average and the distance to the BS less this route can be changed dynamically if the cost function is changed
than the average as a root CH. This root regroups data from other CHs, in the same turn.

13
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

Table 5
Advantages and disadvantages of class 3.
Protocol Strengths Limitations
Equilibrating energy consumption in the network. Electing CH according to the energy, it may be further from the BS.
E-LEACH
CHs selection depending on the remaining energy of nodes. Chooses a root-CH based only on the energy whatever the
distance to the BS.

Extends the network lifetime. No. of nodes in clusters is not equal.

BRE-LEACH CH selection according to the residual energy.
Pick out a parent CH depending on the remaining energy In every turn, the no. of CHs may be is not the same.
and the distance to the BS.

Considers the remaining energy in CHs selection. Uses single hop in intra-clusters.
Defines the maximum number of nodes in each cluster.
EDMHT-LEACH
Divides network into multi levels.
Uses a dynamic route of data.

Table 6
General characteristics of LEACH-based protocols (Class 3).
Protocol Year Research domain Network type Modeling Parameters Complexity
LEACH 2000 Load balancing homogeneous Energy saving Low
E-LEACH 2012 Energy-Efficient homogeneous Remaining energy Yes
BRE-LEACH 2019 Energy-Efficiency homogeneous Remaining energy and distance to the BS Yes
EDMHT-LEACH 2017 Multi-level and dynamic route homogeneous Residual energy and distance Yes

The EDMHT-LEACH saves energy by selecting CH according to the The discussed clustering routing protocols are compared based on
residual energy of nodes. Thus, through this technique, also indepen- different aspects, such as clustering method (centralized or distributed),
dent nodes can transfer their data to the BS. Therefore, this enhanced CH selection method (either random or considering some parame-
approach decreases the total energy consumption and extends the net- ters like residual energy, distance, the density of node, etc.), mobil-
work lifespan. Furthermore, EDMHT-LEACH has some drawbacks such ity of node, scalability, deployment of nodes, communication tech-
as it uses the same process as LEACH for transmitting data for intra- niques (single-hop or multi-hop or both), energy-efficiency, localization
cluster. So, the dynamic routing technique is used just for inter-cluster information.
[78]. A flowchart of the EDMHT-LEACH protocol is given in [74]. The set of the discussed protocols use the TDMA schedule which
allows nodes to be in standby mode while waiting for their data
Table 5 introduces the advantages and disadvantages of protocols
transmission turn. They consider that the sink is fixed in a defined
base on the CH selection and data transmission technique to enhance
place, and then they are based on the data aggregation technique.
the original LEACH algorithm.
Finally, all the discussed protocols use random rotation.
The discussed improved LEACH hierarchical routing protocols based The random selection of CH in the network affects the entire per-
on the CH selection and data transmission techniques to enhance the formance of WSN. Due to the tasks carried out by the CH in collecting,
LEACH algorithm are summarized in Table 6 according to the research aggregating data, and sending them to the sink, the CH selection
domain, network type, modeling parameters of each protocol, and the impacts energy consumption and the life-cycle of WSN. So, the right
complexity of the algorithm. choice of CH node is a challenge in routing protocols. Several LEACH-
based clustering protocols were developed to deal with this issue based
3.4. Comparative analysis and discussion on different aspects. As illustrate in Table 9, numerous of discussed CH
selection-based protocols pick out the CH by considering the residual
energy parameter, such as the LEACH-B and the I-LEACH. LEACH-C
Considering the special features of WSN, the clustering technique uses the centralization technique at the BS and the residual energy to
is the most challenging field in WSN owing to its various advan- select CH and form clusters in the network. Furthermore, LEACH-M
tages, which decrease the number of routed packets and achieve good and LEACH-ME are based on the mobility of nodes and residual en-
energy-efficiency in the network. ergy. VH-LEACH, LEACH-T, and TB-LEACH utilize the energy efficiency
The basic LEACH approach faces numerous restrictions in terms parameter. Then, Improved-LEACH, EADCR, and BN-LEACH consider
of CH selection and data transmission techniques, which impact the distance and residual energy.
network lifetime and increase energy consumption. Another essential problem in designing routing protocols in WSN
This survey classified LEACH variants clustering protocols into CH is data transmission techniques, which have an important impact on
selection-based routing protocols, data transmission based routing pro- the network lifetime. In the single-hop transmission, the CH sends the
tocol, and protocols based on both CH selection and data transmission aggregated data from its cluster members directly to the BS as used
to enhance the LEACH performances. in the basic LEACH protocol. This technique is useful only in small
network areas.
These protocols based on LEACH were proposed to improve the
On the other hand, the multi-hop technique is used to enhance
classical LEACH approach through decreasing energy consumption and
data transmission. In the multi-hop communication, the CH uses some
extending the lifetime of WSNs.
intermediate nodes (either nodes or other CHs) to forward its data to
The discussed LEACH-descendants exhibit better performance than
the BS. The principal goal of the multi-hop technique is to maintain
the original LEACH protocol by considering the CH selection approach the distance between the source and the destination node less than the
and data transmission techniques. threshold distance; because the energy dissipation in the radio model
Table 7 provides the improvements made by LEACH derived proto- is proportional to the distance. If the distance between the source and
cols compared to classical LEACH. the destination nodes is less than the threshold, the energy dissipation
A comparative analysis of LEACH variants has been highlighted as is in the function of 𝑑 2 . On the other side, the energy consumption is
shown in Tables 8 and 9. in the function of 𝑑 4 .

14
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

Table 7
Comparison of LEACH and its descendants.
Protocol Difference from the original LEACH
A-LEACH It aims to improve the stability, energy-saving, and reduce the chance of nodes failure by using CAG nodes.
LEACH-B In each turn, LEACH-B guarantees a near-optimal number of CHs. Then it takes into account the residual energy for selecting CHs after the first turn.
LEACH-C Centralized algorithm, the BS creates clusters and elects CHs according to nodes location and the residual energy of nodes.
LEACH-Cell LEACH-Cell splits the network into cells (sections), and every seven close cells form clusters. Every cell contains a cell-head, and each cluster has a CH.
E-LEACH CH selection based on the remaining energy of nodes, and an optimal CH (with maximum remaining energy) is elected to group other CHs packets, then
direct data directly to the sink
LEACH-F CHs selection and clusters formation are based on the centralizing approach. Thus, clusters are fixed once they are formed at the beginning.
LEACH-M LEACH-M uses mobility of both normal nodes and CHs for transfer data to the BS. Therefore, it is adapted for mobility environments.
LEACH-ME Sensor nodes with minimum mobility compared to its neighbors are selected as CHs.
I-LEACH CHs Election on the basis of the remaining energy, nodes location, and neighbor number. Then, it updates the threshold function by adding the residual
energy. Thus, I-LEACH envisages the optimal number of clusters.
LEACH-TL In LEACH-TL, two kinds of CHs are generated: primary and secondary CH. When the primary CH has remaining energy less than the average or its
distance to BS is more than the average, The secondary CH takes the role of CH and sends aggregated data to the primary CH, whereas this latter
communicates only with the BS.
V-LEACH When a Cluster-Head dies, a vice Cluster-Head replaces it.
VH-LEACH CHs selection on the basis of the remaining energy, and when a cluster head’s residual energy attains a threshold, a VH node takes the role of the CH.
O-LEACH Orphan nodes can join clusters and send their data to the sink.
LEACH-T Fixation of the number of time slots in each CH TDMA. LEACH-T considers the residual energy for establishing clusters, and it envisages an optimal
number of CHs in the network.
TB-LEACH The number of CHs in every epoch is fixed. Thus, CHs selection in TB-LEACH depends on a random timer.
Improved-LEACH The distance and the residual energy are two factors that are considered in selecting CHs.
EADCR Improve the network lifetime by applying the FCM approach to establish clusters, selecting CHs using the fitness function which depends upon the
Euclidean distance and the residual energy of nodes.
BN-LEACH Using the Bayesian Network model to pick out CHs.
W-LEACH In W-LEACH, there are two types of nodes: active and sleeping nodes, that they are attributed by the base station. Thereby, CHs are chosen depending
on the weight value Wi that it is computed based on the residual energy and the density of nodes.
WD-LEACH WD-LEACH divides the network into active and sleeping nodes without using centralization in the BS using the number of neighbor nodes.
L-LEACH L-LEACH is based on the distance to choose CHs. The multi-hop method is used as the single hop depending on the distance from the base station.
BRE-LEACH Only nodes with enough remaining energy can participate in CH selection. BRE-LEACH updates the threshold formula by adding the ratio of residual
energy per initial energy. Thus it elects the root CH with maximum energy and minimum distance to the BS to group data from CHs, and sends it to the
sink.
MH-LEACH In place of forwarding packets to the sink directly, MH-LEACH uses the multi-hop technique between CHs to achieve the sink through constructing a
routing table for each CH.
MHT-LEACH MHT-LEACH divides CHs into two groups according to the distance from the BS. Thus, it uses multi-hop between CHs to reach the BS.
IMHT-LEACH IMHT-LEACH divides CHs into multi-levels, and each level has a length of 𝑑0 ∕2. Each CH sends its data to the BS using the multi-hop technique
according to its own routing table.
DMHT-LEACH CHs are distributed into multi-level, and each level has a length of 𝑑0 ∕2. Each CH creates its routing table according to a cost function. Thus, CH can
change its data route dynamically several times in the same turn once the cost function change .
EDMHT-LEACH EDMHT-LEACH updates the threshold function to choose CHs to depend upon the remaining energy of nodes. It determines the maximum number of
nodes that it can be in a cluster. CHs and INs nodes (Independent nodes) generate their routing table according to a cost function. The route in routing
tales are organized dynamically on the basis of the cost function.

The use of single-hop or multi-hop is depending on the application classical clustering LEACH protocol, such as A-LEACH, LEACH-TL, and
and the network (small or large). For example, if the network is L-LEACH.
small, it is not worth using the multi-hop because the single-hop is
Finally, in the discussed protocols, some protocols improve both the
enough and the multi-hop will consume more energy. On the other
CH selection by using the residual energy parameter and use the multi-
hand, if the network is wide, the multi-hop is necessary to minimize
hop to enhance data transmissions, such as the E-LEACH, BRE-LEACH,
the energy consumption because the single-hop requires more energy
and the EDMHT-LEACH.
which influences the lifetime of the entire network.
This survey provides a comprehensive study of LEACH descendant From this survey and the comparative analysis, some conclusions
clustering protocols, which classify them into protocols that improve will be concluded: designing a good routing protocol faces to two
LEACH by improving the CH selection phase, protocols that enhancing major constraints, selecting a reliable and an efficient CH is necessary,
LEACH through enhancing data communication (single-hop and multi- also using an efficient data communication techniques is important to
hop techniques), and protocols that evolve both CH selection and data extend the network lifetime and reducing the energy consumption.
communication techniques.
Table 10 shows the improvement of the network lifetime and the
Our survey grouped the improved LEACH protocols based on data
energy consumption of LEACH-variant protocols compared to LEACH.
communication into three groups: (1) single-hop-based protocols, such
as LEACH-F, V-LEACH, W-LEACH, and WD-LEACH; (2) multi-hop-based Some numerical results of LEACH-descendant protocols are given in
protocols to enhance LEACH performances, such as LEACH-cell, O- the literature, such as in I-LEACH, VH-LEACH, TB-LEACH, Improved-
LEACH, MH-LEACH, MHT-LEACH, IMHT-LEACH, and DMHT-LEACH; LEACH, BN-LEACH, V-LEAH, W-LEACH, MH-LEACH, L-LEACH,
and (3) protocols that used both routing techniques to improve the E-LEACH, and BRE-LEACH. Thereby, through numerical results given

15
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

Table 8
Characteristics of LEACH-based routing protocols.
Class Category Protocol No. of CHs No. of nodes in cluster Clustering method Localization
LEACH Uncertain Uncertain Distributed No
LEACH-B Certain Uncertain Distributed Yes
Residual energy
I-LEACH Certain Uncertain Distributed No
Centralization LEACH-C Certain Uncertain Centralized Yes

LEACH-M Uncertain Uncertain Distributed Yes

Mobility
CH selection LEACH-ME Uncertain Uncertain Distributed Yes

VH-LEACH Uncertain Uncertain Distributed Yes

Energy efficiency LEACH-T Certain Uncertain Distributed No
TB-LEACH Certain Uncertain Distributed No

Improved-LEACH Uncertain Uncertain Distributed Yes

Distance EADCR Certain Uncertain Centralized Yes
BN-LEACH Uncertain Uncertain Distributed Yes

LEACH-F Certain Uncertain Centralized Yes

V-LEACH Uncertain Uncertain Distributed Yes
Single Hop
W-LEACH Uncertain Uncertain Centralized Yes
WD-LEACH Uncertain Uncertain Distributed Yes

LEACH-Cell Uncertain Uncertain Distributed Yes

Data transmission O-LEACH Uncertain Uncertain Distributed No
MH-LEACH Uncertain Uncertain Distributed Yes
Multi Hop
MHT-LEACH Uncertain Uncertain Distributed Yes
IMHT-LEACH Uncertain Uncertain Distributed Yes
DMHT-LEACH Uncertain Uncertain Distributed Yes

A-LEACH Uncertain Uncertain Distributed No

Single/Multi Hop LEACH-TL Uncertain Uncertain Distributed Yes
L-LEACH Uncertain Uncertain Distributed Yes

E-LEACH Certain Uncertain Distributed No

CH selection and Data transmission BRE-LEACH Uncertain Uncertain Distributed Yes
EDMHT-LEACH Uncertain Certain Distributed Yes

in the literature, we have calculated the percentage of the network • In terms of the delay of the FND (first node death), the BN-LEACH
lifetime of some protocols, such as LEACH-B, VH-LEACH, LEACH- performs better by 19.5%, 46.06%, and 52.61% than LEACH-C,
T, LEACH-TL, MHT-LEACH, IMHT-LEACH, and EDMHT-LEACH. On WEEC, and LEACH, respectively. In terms of the delay of the
LND (last node death), the BN-LEACH performs better by 23.54%,
the other hand, due to a lack of numbers, we cannot compute the
28.32%, and 43.55% than WEEC, LEACH-C, and LEACH.
percentage of the improvement of network lifetime and the energy
• In [62], there are two scenarios, in the 1st scenario (50 nodes,
consumption of LEACH-C, LEACH-M, LEACH-ME, EADCR, LEACH-F, area of 50X50 m2 , and BS(25,100)), the MH-LEACH decreases
WD-LEACH, LEACH-cell, O-LEACH, DMHT-LEACH, and A-LEACH. energy consumption by 9%, and increases the network lifetime by
Another comparison between protocols is given as follows: 4.57% compared to LEACH. In the 2nd scenario (100 nodes, area
of 70X70 m2 , and BS(35,140)), the MH-LEACH reduces energy
• In [34], the network lifetime is increased with I-LEACH by 59% consumption by 6% and enhances the network lifetime by 4.61%
than the DBS approach, and as well as 29% than LEACH-C proto- compared to LEACH.
col. Thus, the energy consumption in I-LEACH has been decreased A complete comparison among the current survey and previous
by 55% than DSB, and by 27% than the LEACH-C algorithm. In surveys in [12–16] is given in Table 11.
addition to the number of data packets that are forwarded to
the BS has been raised to 56% using the I-LEACH approach in 4. Future works and research challenges
comparison to LEACH, 21% than LEACH-C, and 46% than DSB.
The comparative analysis and discussions exhibit that both CH
• LEACH-C forwards about 40% more data per unit energy than
selection and data transmission techniques are the major challenges
LEACH [36]. in designing routing protocols in WSN. All the discussed clustering
• In the LEACH-ME, the successful communication rate is 16% protocols associated with LEACH focused on reducing energy consump-
at the mobility factor of 4.0, which is better than the LEACH- tion and maximizing the network lifetime. For this reason, various
M. For high mobility (mobility factor 4 and above), the over- approaches have been proposed to improve the basic LEACH protocol.
head of LEACH-ME is 22% more than that of the overhead of A variety of parameters have been considered by these approaches. But,
there is nonetheless a wide variety of open research problems to be
LEACH-M [41].
addressed in the future.
• The LEACH-VH protocol consumes 26.69% of the total energy,
The main objectives for developing LEACH-derived protocols for
which is about 41.178%, 27.31%, 18.418%, 17.088%, 13.596%, WSNs are: enhancing the WSN performance, energy-efficiency, energy
and 11.578% less than LEACH, LEACH-C, LEACH-F, LEACH-L, distribution between nodes, enhancing the scalability, selecting the best
TL-LEACH, and V-LEACH, respectively [44]. node to be the CH in each cluster, increasing the security in WSN,
• Some percentage comparison between the EADCR and other pro- reduction of network delay, increasing the stability of the network.
tocols like FCM, REHR is provided in [49]. This survey proposes some future issues as follows:

16
 I. Daanoune et al.
Table 9
Comparison based on different metrics.
Class Category Protocol CH selection Communication Mobility Scalability Energy Residual Deployment
method Efficiency energy
LEACH Random Single Hop Static Limited High No Random

LEACH-B Random, Residual Energy Single Hop Static Good Very High Yes Random
Residual energy
I-LEACH Residual Energy Single Hop Static Very Good Very High Yes Random
Centralization LEACH-C Residual Energy Single Hop Static Good Very High Yes Random

LEACH-M Residual Energy, Mobility Single Hop Mobile Very Good Very High Yes Random
Mobility
CH selection LEACH-ME Residual Energy, Mobility Single Hop Mobile Limited High Yes Random

VH-LEACH Residual energy Single Hop Static Very Good Very High Yes Random
Energy efficiency LEACH-T Residual Energy Single Hop Static Good High Yes Random
TB-LEACH Random Single Hop Static Limited High No Random

Improved-LEACH Distance, Residual Energy Single Hop Static Good High Yes Random
Distance EADCR Euclidean Distance, Residual Single hop/Multi-hop Static Good Very High Yes Random
Energy
17

BN-LEACH Density, distance residual energy Single hop Static Good High Yes Random

LEACH-F Random Single Hop Static Limited Very High No Random

V-LEACH Random Single Hop Static Good Very High No Random
Single Hop
W-LEACH Random Single Hop Static Good High No Random
WD-LEACH Random Single Hop Static Good High No Random

LEACH-Cell Random Multi-Hop Static Very Good Very High No Random

Data transmission O-LEACH Random Multi-Hop Static Good High No Random
MH-LEACH Random Multi-Hop Static Good Very High No Random
Multi Hop
MHT-LEACH Random Multi-Hop Static Good Very High No Random
IMHT-LEACH Random Multi-Hop Static Good Very High No Random
DMHT-LEACH Random Multi-Hop Static Good Very High No Random

A-LEACH Random Static Good Very High No Random

Single/Multi Hop LEACH-TL Random Single Hop/Multi-Hop Static Very Good Very High No Random
L-LEACH Random Static Very Good Very High No Random

E-LEACH Residual Energy Multi-Hop Static Very Good Very High Yes Random

Ad Hoc Networks 114 (2021) 102409

CH Selection and Data transmission BRE-LEACH Residual energy Multi-Hop Static Very Good Very High Yes Random
EDMHT-LEACH Residual energy Multi-Hop Static Good Very High Yes Random
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

• Complexity: Owing to the composition of the sensor node, which Table 10

is characterized by a restricted calculation capability, and small Improvement of the network lifetime and energy consumption of protocols compared
to LEACH.
memory. So, the complexity influences the global protocol per-
Protocol Network lifetime Energy consumption
formance in WSNs. Consequently, scheming a clustering routing
protocol with low complexity is very crucial. It must that the clus- LEACH-B 25% –
I-LEACH 67% 62%
tering routing protocol takes into consideration many parameters
VH-LEACH 59.04% 41.178%
with the lowest complexity to minimize the energy consumption LEACH-T 60% –
in the network. TB-LEACH 20 to 30% –
• Deployment: Deployment is one of the major problems in WSN Improved-LEACH 25% –
due to it presents an important parameter to determine the net- BN-LEACH 52.61% –
work performance. Some routing protocols improve the LEACH V-LEACH 49.37% –
W-LEACH 60% –
algorithm without considering the deployment of nodes in the
MH-LEACH 4.57% 9%
network. In some cases, the deployment of nodes increases the MHT-LEACH 77.70% –
energy consumption in the entire network. Therefore, the deploy- IMHT-LEACH 55.31% –
ment of nodes in the network must be made by considering some LEACH-TL 52.76% –
metrics, such as the coverage, energy, connectivity, and so forth. L-LEACH 51% –
E-LEACH 49% –
• Fault tolerance: Fault tolerance is another important problem in
BRE-LEACH 55.73% 55.73%
WSN because it affects the network lifetime. In WSN, nodes can be EDMHT-LEACH 88.43% –
deployed in a surrounding environment that cannot be accessed
by humans and keeps unattended. So the component failure of a
node is inevitable. Especially, if the CH breaks down, it results in
more damage to the overall network because the CH is respon- In future work, we will propose a clustering routing protocol that will
sible to route its cluster members data to the BS. Consequently, address as many issues as possible. It will take into account several
implementing a fault-tolerance parameter in clustering routing parameters of the QoS, more security, tolerates failures, and especially
protocols is important to avoid any fault caused by link breaks, in the case of death CH with the least complexity. It will also take into
death of a node, and especially how to re-group and select another account the deployment of nodes and will not be deployed randomly
CH in case of CH failure. as the protocols discussed in this survey. Finally, our future work will
• Modulation techniques: The choice of modulation and channel take into account modulation and coding techniques to improve the
encoding with optimal parameters help to reduce energy con- efficiency, reliability, and lifetime of the network.
sumption and ensure greater safety in WSNs. In this context, an
optimal modulation with an adapted channel coding method is
an extremely crucial technique in the physical layer in WSN. So, 5. Conclusion
implementation of modulation and coding techniques in LEACH-
descendant protocols are future opportunities for researchers. Numerous routing protocols based on the clustering technique have
• QoS: Different clustering protocols of WSN have been considered been suggested owing to its advantages. These advantages like data
energy-efficiency as a significant key to improving the network aggregation, load balancing, less energy consumption, and scalability.
scalability and the network lifetime. Nonetheless, some recent LEACH is treated as the first classical energy-efficient protocol using the
applications that use the WSNs have diverse constraints, which clustering routing. It was proposed to raise the lifetime and decrease
require QoS (Quality of Service) metrics in terms of end to
the consumption of energy in the network. Nevertheless, LEACH has
end delay, reliability, bandwidth, scalability, latency, adaptiv-
several drawbacks in connecting with energy efficiency, CH selection,
ity, throughput, robustness, etc [79]. Therefore, researchers have
and data transmission, which limited the network life span. To sur-
the opportunity to propose and develop clustering-based energy
mount these issues, many routing protocols have been suggested by the
efficiency protocols based on the QoS (reliability, bandwidth,
scalability, latency, adaptivity, throughput, robustness, and so researcher to enhance the conventional LEACH protocol. In this survey,
forth) to address efficiency issues in WSNs. the original LEACH approach has been discussed with its benefits and
• Meta-heuristic: Meta-heuristic (or heuristic) are nature-inspired drawbacks. Furthermore, based on the drawbacks of LEACH, we have
algorithms that have been used to develop solutions to determin- surveyed 27 LEACH descendants routing protocols. These discussed
istic combinatorial optimization issues in WSN. These approaches LEACH-based protocols are given in a new taxonomy, which depends
have been proposed to solve clustering problems, such as the on the CH selection methods and the data transmission techniques.
selection of CH, the optimal amount of CH in the network, The new taxonomy classified these protocols into three classes: (1)
deployment of nodes, data aggregation, and localization [80]. protocols that improve LEACH by improving the CH selection phase;
Meta-heuristic algorithms are important open research challenges (2) protocols that enhancing LEACH through enhancing the data trans-
for improving LEACH-descendant protocols. There are several im- mission techniques, and protocols that evolve both CH selection and
provements that can be adopted in the meta-heuristic algorithms data communication techniques. Besides, to make this classification
in order to implement them in the LEACH variant protocols to more sunny and clear, we divided each class into categories, such as
enhance the network lifetime. Future researchers may develop residual energy, energy efficiency, single-hop, multi-hop, and so forth.
hybrid protocols characterized by the benefits of several meta-
These metrics are used to organize the discussed protocols owing to
heuristics. These hybrid protocols take into account the majority
their impact on the stability and the lifetime of the WSN. This research
of QoS, adaptive modulation coding techniques, etc to improve
discussed 27 clustering algorithms with the LEACH approach. The ben-
WSN performance.
efits and drawbacks of each discussed protocol are given in this work,
• Security: Encryption security in clustering protocols based on
LEACH present a major constraint in WSN due to the nature also the difference from the original LEACH. Furthermore, to evaluate
of the sensor node, which has a small memory and a restricted and compare these LEACH descendants protocols, this survey examined
calculation capability. Owing to the important tasks performed many metrics, such as the clustering method, self-organization, the
by the CH such as aggregating data and transmitting it to the number of CHs, the number of nodes in clusters, energy-efficiency,
BS, it represents the device most accessible by hackers. Therefore, scalability, the communication technique, and so forth. The objective of
authentication and encryption algorithms should be applied at the this survey is to classify the improved LEACH protocols to facilitate the
CH level to prevent attacks. choice of efficient protocols for researchers according to their needs.

18
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

Table 11
Comparison among this survey and the existing surveys.
Current Survey [12] [13] [14] [15] [16]
Classification of protocols Yes No No No No Yes

Well detailed taxonomy Yes No No No No No

Detailed description of distinct Yes No No No No No

protocols

Survey range CH selection and Global presentation CHs selection Communication Overview of LEACH Election of CHs and
data transmission of LEACH and other protocols techniques protocols and its improved cluster formation
protocols hierarchical routing protocols routing protocols
protocols

No. of parameters used to 17 None 5 None 4 6

compare protocols

Advantages and disadvantages of Yes No No Yes No Yes

each protocol

Comparison with preceding Yes No No No No Yes

reviews

No. of discussed protocols 27 6 12 10 12 21

Acknowledgment [14] M. Mahmuddin, W.A. Alabadleh, L.M. Kamarudin, A comparative study on

hoping mechanism of LEACH protocol in Wireless Sensor Netw.: A survey, IOP
Conf. Ser. Mater. Sci. Eng. 551 (2019) 012056.
The authors would like to gratefully thank the reviewers and editor
[15] Rishikesh, Abhijeet Das, Parma Nand Astya, A relative survey of various LEACH
of Ad Hoc Networks for their support and their suggestion that served
based routing protocols in wireless sensor networks, in: 2017 International
to improve the quality of this paper. Conference on Computing, Communication and Automation, ICCCA, IEEE, 2017,
pp. 630–636.
Declaration of competing interest [16] Mohammed Al-Shalabi, Mohammed Anbar, Tat-Chee Wan, Ahmad Khasawneh,
Variants of the low-energy adaptive clustering hierarchy protocol: Survey, issues
The authors declare that they have no known competing finan- and challenges, Electronics 7 (8) (2018) 136.
[17] Mohammed Abdulkarem, Khairulmizam Samsudin, Fakhrul Zaman Rokhani,
cial interests or personal relationships that could have appeared to
Mohd Fadlee A. Rasid, Wireless sensor network for structural health monitoring:
influence the work reported in this paper. A contemporary review of technologies, challenges, and future direction, Struct.
Health Monit. (2019) 147592171985452.
References [18] A.J. Watt, M.R. Phillips, C.E-A. Campbell, I. Wells, S. Hole, Wireless Sensor Netw.
for monitoring underwater sediment transport, Sci. Total Environ. 667 (2019)
[1] Mandeep Kaur, Amit Munjal, Data aggregation algorithms for wireless sensor 160–165.
network: A review, Ad Hoc Netw. 100 (2020) 102083. [19] Mahmoud Badawy, Mona Moustafa, Hesham Arafat Ali, Yousry Abdulazeem, A
[2] Zaher Al Aghbari, Ahmed M. Khedr, Walid Osamy, Ifra Arif, Dharma P. Agrawal, review on cluster-based routing for wireless sensor network, Int. Res. J. Eng.
Routing in Wireless Sensor Netw. using optimization techniques: A survey, Wirel. Technol. 06 (12) (2019) 16.
Pers. Commun. (2019). [20] Ikram Daanoune, Abdennaceur Baghdad, Abdelhakim Ballouk, An enhanced
[3] Noman Shabbir, Syed Rizwan Hassan, Routing protocols for Wireless Sensor energy-efficient routing protocol for wireless sensor network, Int. J. Electr.
Netw. (WSNs), in: Wireless Sensor Netw. - Insights and Innovations, 2017. Comput. Eng. 10 (5) (2020) 8.
[4] R. Vinodhini, C. Gomathy, MOMHR: A dynamic multi-hop routing protocol for [21] Tarunpreet Kaur, Dilip Kumar, A survey on QoS mechanisms in WSN for
WSN using heuristic based multi-objective function, Wirel. Pers. Commun. 111 computational intelligence based routing protocols, Wirel. Netw. (2019).
(2) (2020) 883–907. [22] Priyanka Dubey, Dr.S. Veenadhar, Dr. Sanjeev Gupta, Survey on energy efficient
[5] Nikolaos A. Pantazis, Stefanos A. Nikolidakis, Dimitrios D. Vergados, Energy- clustering and routing protocols of wireless sensor network, Int. J. Sci. Res. 5
efficient routing protocols in Wireless Sensor Netw.: A survey, IEEE Commun. (1) (2019) 7.
Surv. Tutor. 15 (2013) 551–591. [23] Sadeer Rasheed Ahmed, Mohammed Aboud Kadhim, Tarek Abdulkarim, Wireless
[6] Maryam Shafiq, Humaira Ashraf, Ata Ullah, Shireen Tahira, Systematic literature Sensor Netw. improvement using LEACH algorithm, IOP Conf. Ser. Mater. Sci.
review on energy efficient routing schemes in WSN – A survey, Mob. Netw. Appl. Eng. 518 (2019) 052023.
(2020). [24] Mohd Nazim Jambli, Mohd Imran Bandan, Kosheila Sundram Pillay, Sinar-
[7] Biswa Mohan Sahoo, Tarachand Amgoth, Hari Mohan Pandey, Particle swarm
wati Mohamad Suhaili, An analytical study of LEACH routing protocol for
optimization based energy efficient clustering and sink mobility in heterogeneous
wireless sensor network, in: 2018 IEEE Conference on Wireless Sensors, ICWiSe,
wireless sensor network, Ad Hoc Netw. 106 (2020) 102237.
IEEE, 2018, pp. 44–49.
[8] Ramesh Rajagopalan, Pramod Varshney, Data-aggregation techniques in sensor
[25] Ikram Daanoune, Abdennaceur Baghdad, Abdelhakim Balllouk, BRE-LEACH: A
networks: a survey, IEEE Commun. Surv. Tutor. 8 (4) (2006) 48–63.
new approach to extend the lifetime of wireless sensor network, in: 2019 Third
[9] Xuxun Liu, A survey on clustering routing protocols in Wireless Sensor Netw.,
International Conference on Intelligent Computing in Data Sciences, ICDS, IEEE,
Sensors 12 (8) (2012) 11113–11153.
2019, pp. 1–6.
[10] W.R. Heinzelman, A. Chandrakasan, H. Balakrishnan, Energy-efficient commu-
[26] Chunyao Fu, Zhifang Jiang, Wei Wei, Ang Wei, An energy balanced algorithm
nication protocol for wireless microsensor networks, in: Proceedings of the
of LEACH protocol in WSN, Int. J. Comput. Sci. Issues 10 (1) (2013) 6.
33rd Annual Hawaii International Conference on System Sciences, vol. 1, IEEE
Comput. Soc., Maui, HI, USA, 2000, p. 10. [27] Wenliang Wu, Naixue Xiong, Chunxue Wu, Improved clustering algorithm based
[11] Ashish Khanna, Deepak Gupta, Siddhartha Bhattacharyya, Vaclav Snasel, Jan on energy consumption in wireless sensor networks, IET Netw. 6 (3) (2017)
Platos, Aboul Ella Hassanien (Eds.), International Conference on Innovative 47–53.
Computing and Communications: Proceedings of ICICC 2019, vol. 2, in: Advances [28] https://ch.mathworks.com/matlabcentral/fileexchange/66574-leach, Last
in Intelligent Systems and Computing, vol. 1059, Springer Singapore, Singapore, visit{October 2020}.
2020. [29] Ramin Yarinezhad, Seyed Naser Hashemi, Increasing the lifetime of sensor net-
[12] Vishal Kumar Arora (Research Scholar), Vishal Sharma, Monika Sachdeva, A works by a data dissemination model based on a new approximation algorithm,
survey on LEACH and other’s routing protocols in wireless sensor network, Optik Ad Hoc Netw. 100 (2020) 102084.
127 (16) (2016) 6590–6600. [30] Mu Tong, Minghao Tang, LEACH-B: An improved LEACH protocol for wireless
[13] Adnan Yousaf, Faiza Ahmad, Shahzaib Hamid, Fahad Khan, Performance com- sensor network, in: 2010 International Conference on Computational Intelligence
parison of various LEACH protocols in Wireless Sensor Netw., in: 2019 IEEE 15th and Software Engineering, IEEE, 2010, pp. 1–4.
International Colloquium on Signal Processing & Its Applications, CSPA, IEEE, [31] Amandeep Kaur, Amit Grover, LEACH and extended LEACH protocols in wireless
2019, pp. 108–113. sensor network-A survey, Int. J. Comput. Appl. 116 (10) (2015) 1–5.

19
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

[32] J. Gnanambigai, Dr.N. Rengarajan, K. Anbukkarasi, Leach and its descendant [60] Rajat Kandpal, Rajesh Singh, H.L. Mandoria, Analysis on enhancements in LEACH
protocols: A survey, Int. J. Commun. Comput. Technol. 01 (3) (2012). protocol for WSN, Int. J. Comput. Sci. Eng. 2 (9) (2015) 8.
[33] Raed M. Bani Hani, Abdalraheem A. Ijjeh, A survey on LEACH-based energy [61] Wassim Jerbi, Abderrahmen Guermazi, Hafedh Trabelsi, O-LEACH of routing
aware protocols for Wireless Sensor Netw., J. Commun. 8 (3) (2013) 192–206. protocol for Wireless Sensor Netw., in: 2016 13th International Conference on
[34] Z. Beiranvand, A. Patooghy, M. Fazeli, I-LEACH: An efficient routing algorithm to Computer Graphics, Imaging and Visualization, CGiV, IEEE, 2016, pp. 399–404.
improve performance & to reduce energy consumption in Wireless Sensor Netw., [62] Jose Henrique Brandao Neto, Antoniel da Silva Rego, Andre Ribeiro Cardoso,
may, 2013. Joaquim Celestino Jr., MH-LEACH: A distributed algorithm for multi-hop com-
[35] Wu Xinhua, Wang Sheng, Performance comparison of LEACH and LEACH-
munication in Wireless Sensor Netw., in: ICN 2014 : The Thirteenth International
c protocols by NS2, in: 2010 Ninth International Symposium on Distributed
Conference on Netw., 2014, pp. 55–61.
Computing and Applications to Business, Engineering and Science, IEEE, 2010,
[63] http://www.njavaid.com/downloads.aspx, Last visit{October 2020}.
pp. 254–258.
[36] W.B. Heinzelman, A.P. Chandrakasan, H. Balakrishnan, An application-specific [64] Emad Alnawafa, Ion Marghescu, MHT: Multi-hop technique for the improvement
protocol architecture for wireless microsensor networks, IEEE Trans. Wireless of leach protocol, in: 2016 15th RoEduNet Conference: Networking in Education
Commun. 1 (4) (2002) 660–670. and Research, IEEE, 2016, pp. 1–5.
[37] Yu Xiuwu, Liu Qin, Liu Yong, Hu Mufang, Zhang Ke, Xiao Renrong, Uneven [65] Jian-hua Huang, Zi-ming Zhao, Yu-bo Yuan, Ya-dong Hong, Multi-factor and
clustering routing algorithm based on glowworm swarm optimization, Ad Hoc distributed clustering routing protocol in Wireless Sensor Netw., Wirel. Pers.
Netw. 93 (2019) 101923. Commun. 95 (3) (2017) 2127–2142.
[38] Siva D. Muruganathan, Daniel C.F. Ma, Rolly I. Bhasin, Abraham O. Fapojuwo, A [66] Emad Alnawafa, Ion Marghescu, IMHT: Improved MHT-LEACH protocol for
centralized energy-efficient routing protocol for wireless sensor networks, IEEE wireless sensor networks, in: 2017 8th International Conference on Information
Commun., 43(3), http://dx.doi.org/10.1109/MCOM.2005.1404592. and Communication Systems, ICICS, IEEE, 2017, pp. 246–251.
[39] Ricardo Silva, Jorge Sa Silva, Fernando Boavida, Mobility in wireless sensor [67] Brijesh Iyer, S.L. Nalbalwar, Nagendra Prasad Pathak (Eds.), Computing, Com-
networks – Survey and proposal, Comput. Commun. 52 (2014) 1–20. munication and Signal Processing: Proceedings of ICCASP 2018, in: Advances in
[40] Do-Seong Kim, Yeong-Jee Chung, Self-organization routing protocol supporting Intelligent Systems and Computing, vol. 810, Springer Singapore, 2019.
mobile nodes for wireless sensor network, in: First International Multi-
[68] Mohanad H. Wasmi, Salah A. Aliesawi, Wesam M. Jasim, Distributed semi-
Symposiums on Computer and Computational Sciences, IMSCCS’06, IEEE, 2006,
clustering protocol for large-scale wireless sensor networks, Int. J. Eng. (2018)
pp. 622–626.
8.
[41] G. Santhosh Kumar, Paul M.V. Vinu, K. Poulose Jacob, Mobility metric based
[69] Jianhua Huang, Danwei Ruan, Yadong Hong, Ziming Zhao, Hong Zheng, IMHRP:
LEACH-mobile protocol, in: 2008 16th International Conference on Advanced
Improved multi-hop routing protocol for wireless sensor netw., J. Phys. Conf. Ser.
Computing and Communications, IEEE, 2008, pp. 248–253.
910 (2017) 012054.
[42] Rajendra Prasad Mahapatra, Rakesh Kumar Yadav, Descendant of LEACH based
routing protocols in Wireless Sensor Netw., Procedia Comput. Sci. 57 (2015) [70] Hana Rhim, Karim Tamine, Ryma Abassi, Damien Sauveron, Sihem Guemara,
1005–1014. A multi-hop graph-based approach for an energy-efficient routing protocol in
[43] http://en.pudn.com/Download/tag/id/17499.html, Last visit{October 2020}. wireless sensor networks, Human-Centric Comput. Inform. Sci. 8 (1) (2018) 1–21.
[44] A. Mehmood, J. Lloret, M. Noman, H. Song, Improvement of the wireless sensor [71] Emad Alnawafa, Ion Marghescu, DMHT-LEACH: Dynamic multi-hop technique for
network lifetime using LEACH with vice-cluster head, Ad-Hoc. Sens. Wirel. Netw. wireless sensor networks, in: 2017 International Symposium on Signals, Circuits
(2015) 1–17. and Systems, ISSCS, 2017, pp. 1–4.
[45] Layla Aziz, A new enhanced version of vleach protocol using a smart path [72] Zheng-Yang Ai, Yu-Tong Zhou, Fei Song, A smart collaborative routing protocol
selection, Int. J. GEOMATE 12 (30) (2017). for reliable data diffusion in IoT scenarios, Sensors 18 (6) (2018) 1926.
[46] Jin Gang Cao, An improvement routing protocol based LEACH for Wireless [73] M. Aishwarya, S. Kirthiga, Relay assisted cooperative communication for wire-
Sensor Netw., Appl. Mech. Mater. 614 (2014) 472–475. less sensor netw., in: 2018 Second International Conference on Advances in
[47] Hu Junping, Jin Yuhui, Dou Liang, A time-based cluster-head selection algorithm Electronics, Computers and Communications, ICAECC, IEEE, 2018, pp. 1–6.
for LEACH, in: 2008 IEEE Symposium on Computers and Communications, IEEE,
[74] Emad Alnawafa, Ion Marghescu, EDMHT-LEACH: Enhancing the performance of
Marrakech, 2008, pp. 1172–1176.
the DMHT-LEACH protocol for wireless sensor networks, in: 2017 16th RoEduNet
[48] Krishnakumar Amirthalingam, Anuratha, Improved LEACH: A modified LEACH
Conference: Networking in Education and Research, RoEduNet, IEEE, 2017, pp.
for wireless sensor network, in: 2016 IEEE International Conference on Advances
1–6.
in Computer Applications, ICACA, IEEE, Coimbatore, India, 2016, pp. 255–258.
[49] Akhilesh Panchal, Rajat Kumar Singh, EADCR: Energy aware distance based [75] Ezzati Abdellah, Said Benalla, Abderrahim BENI Hssane, Moulay Lahcen, Ad-
cluster head selection and routing protocol for Wireless Sensor Netw., J. Circuits vanced low energy adaptive clustering hierarchy, Int. J. Comput. Sci. Eng. 02
Syst. Comput. (2020) 2150063. (07) (2010) 8.
[50] Heydar Ghasemzadeh, Mehdi Rezaeian, Fatemeh Dehghan Touranposhti, Mo- [76] Qian Liao, Hao Zhu, An energy balanced clustering algorithm based on LEACH
hammad Mohsen Ghasemian, BN-LEACH: An improvement on LEACH protocol protocol, in: Proceedings of the 2nd International Conference On Systems
using Bayesian networks for energy consumption reduction in wireless sensor Engineering and Modeling, 2013, pp. 6.
networks, in: 7’Th International Symposium on Telecommunications, IST’2014, [77] Jia Xu, Ning Jin, Xizhong Lou, Ting Peng, Qian Zhou, Yanmin Chen, Improve-
IEEE, Tehran, 2014, pp. 1138–1143. ment of LEACH protocol for WSN, in: 9th International Conference on Fuzzy
[51] Meenakshi Yadav, Bhola Anoop, C.K. Jha, Low –energy adaptive clustering Systems and Knowledge Discovery, 2012, pp. 4.
hierarchy using virtual gird method, Int. J. Adv. Res. Comput. Sci. 8 (8) (2017) [78] Emad Alnawafa, Ion Marghescu, New energy efficient multi-hop routing tech-
246–249. niques for Wireless Sensor Netw.: Static and dynamic techniques, Sensors 18 (6)
[52] P. Manimala, R. Senthamil, A survey on leach-energy based routing protocol, (2018) 1863.
Int. J. Emerg. Technol. Adv. Eng. 3 (2013) pp.657–660. [79] Ikram Daanoune, Abdennaceur Baghdad, Abdelhakim Balllouk, A comparative
[53] Kamaljot Singh, WSN LEACH based protocols: A structural analysis, in: 2015 study between ACO-based protocols and PSO-based protocols in WSN, in: 2019
International Conference and Workshop on Computing and Communication, 7th Mediterranean Congress of Telecommunications, CMT, 2019, pp. 1–4.
IEMCON, IEEE, 2015, pp. 1–7.
[80] Sunil Kumar Singh, Prabhat Kumar, Jyoti Prakash Singh, A survey on successors
[54] A. Ahlawat, V. Malik, An extended vice-cluster selection approach to improve
of LEACH protocol, IEEE Access 5 (2017) 4298–4328.
v leach protocol in WSN, in: 2013 Third International Conference on Advanced
Computing and Communication Technologies, ACCT, IEEE, 2013, pp. 236–240.
[55] Hanady M. Abdulsalam, Layla K. Kamel, W-LEACH: Weighted low energy
adaptive clustering hierarchy aggregation algorithm for data streams in wireless
Ikram Daanoune is a Ph.D. student in Computer Sci-
sensor netw., in: 2010 IEEE International Conference on Data Mining Workshops,
ences and Telecommunications, from Electronics, Energy,
IEEE, 2010, pp. 1–8.
Automatic and Information Processing (EEA&TI) laboratory,
[56] Hanady M. Abdulsalam, Bader A. Ali, W-LEACH based dynamic adaptive data
Hassan II University, Faculty of Sciences & Techniques
aggregation algorithm for Wireless Sensor Netw., Int. J. Distrib. Sens. Netw. 9
(FST), Mohammedia, Morocco. She earned her engineering
(9) (2013) 289527.
[57] Abdelhalim Hnini, Abdellah Ezzati, Mohammed Fihri, Abdelmajid Hajami, W- diploma on electrical & telecommunications at FST Mo-
LEACH decentralized: Weighted LEACH with a decentralized way, IOSR J. hammedia, Hassan II University, in 2018. She received her
Comput. Eng. 16 (2) (2014) 32–39. license degree in electronic engineering and telecommunica-
[58] Abdelhalim Hnini, Mohammed Fihri, Abdellah Ezzati, Mohamed Hanini, W-leach- tions, from FST, Sultan Moulay Slimane University, in 2015.
mobile: an enhanced mobile protocol with a decentralized way in wireless sensor Her areas of interest include wireless sensor networks, ad
network, 2005, p. 10. hoc networks, and routing protocols. She authored many
[59] Arezoo Yektaparast, Fatemeh-Hoda Nabavi, Adel Sarmast, An Improvement on research documents in indexed journals and conferences.
LEACH Protocol (Cell-LEACH), in: 14th International Conference on Advanced
Communication Technology, 2012, pp. 5.

20
I. Daanoune et al. Ad Hoc Networks 114 (2021) 102409

Baghdad Abdennaceur is a University professor at FST

Mohammedia, Hassan II University Casablanca-Morocco. He Abdelhakim Ballouk is a University professor at FST Mo-
has more than 30 years of teaching experience. He received hammedia, Hassan II University Casablanca-Morocco where
his doctorate in Electronics from Lille University France in he educates regulation & digital control, and linear &
1992. He is a member of the EEA&TI laboratory at Hassan II non-linear automatic. He earned his doctorate in Elec-
University. The research of BAGHDAD is focused on optical tronics from Bordeaux University France in 1992, in the
communication, electronic engineering, embedded systems, automatic domain. He is a participant of the EEA&TI
telecommunication systems, information technology, and laboratory at Hassan II University. He oversees and he
antenna. He has an author of numerous research documents co-supervises many doctoral theses. His researches concen-
in indexed journals and conferences. He is a reviewer trated on Automatic (robust commands, predictive controls,
and a member of the organizing committees of several etc.). His researches focused also in Telecommunication
international congresses in the same research area. systems (Antenna, wireless sensor network).

21